polymers

Review
The Chemistry and Physics of Bayfol® HX Film
Holographic Photopolymer
Friedrich-Karl Bruder *, Thomas Fäcke and Thomas Rölle
Covestro Deutschland AG, Building D201, D-51365 Leverkusen, Germany; thomas.faecke@covestro.com (T.F.);
thomas.roelle@covestro.com (T.R.)
* Correspondence: friedrich-karl.bruder@covestro.com; Tel.: +49-214-6009-6205

Received: 7 August 2017; Accepted: 22 September 2017; Published: 26 September 2017

Abstract: Holographic photopolymers are a new technology to create passive diffractive optical elements
by a pure laser interference recording. In this review, we explain the chemistry concepts of light harvesting
in an interference pattern and the subsequent grating formation as chemical response. Using the example
of the newly developed Bayfol® HX film we discuss the reaction-diffusion driven photo-polymerization
process for an index modulation formation to create volume phase gratings. Further we elucidate the
selection of monomer chemistry and discuss details of the recording conditions based on the concept
of exposure dosage and exposure time. Influences ranging from high dosage recording to low power
recording are explained and how to affect the desired diffraction efficiency. Finally, we outline and
demonstrate the process to mass manufacturing of volume phase gratings.

Keywords: photopolymer; volume holographic optical element; photochemistry; interference

recording; mass manufacturing

1. Introduction
In human history, the ability to control light had a fundamental impact on how we interact and
work with each other and with our environment. From the control of the fire to Edison’s light bulb
to modern light emitting (laser) diodes the creation of light and managing it is fundamental to our
life. For a long time, passive optical elements were based only on refraction and reflection. Diffractive
optical elements remained in niche applications like scientific instruments. This is due to the fact that
manufacturing is labor intense and their fragility prevented a wider use in mainstream applications.
We describe here the technology of holographic photopolymers that enable a much wider and
more flexible use to manage light by engineering specific volume holographic optical elements (vHOEs).
Those materials, specifically when they are made available in film format, push forward the boundaries
that diffractive optics exhibited in the past. Photopolymer in general are materials that are sensitized to
react upon light with a chemical response. Holographic photopolymers record an interference pattern
thus exhibits chemical formation of diffractive gratings. A specific benefit to the ones described herein
are those that do not require any chemical and thermal post processing, which is of specific benefit
for the creation of optical components. Mass manufacturing becomes easier and precise and drives
forward the adoption in the consumer space. Their chemical basis as plastic products links them to
classic plastic processing to reach out into popular post hologram recording conversion techniques.
The basic chemical concept, the mechanics of the diffractive grating formation as chemical response
in those new products made available as Bayfol® HX films (Covestro AG, Leverkusen, Germany) is
discussed. Key aspects of influences on the grating formation as function of the spatial frequency (SF),
the exposure dosage and the exposure power are explained. We discuss the photochemistry that is used
and how this influences the photon yield. After the initial chemical response we describe the influence
of the photo-polymerization on the formation of index modulation of the phase grating including
proper selection of polymerizing monomers (usually called writing monomers).

Polymers 2017, 9, 472; doi:10.3390/polym9100472 www.mdpi.com/journal/polymers

Polymers 2017, 9, 472 2 of 35

Holographic recording is done with an interference setup, where a coherent laser light source
is split and illuminates from two angles the photopolymer film which is the recording medium.
This concept is universal and can be used for individual holographic optical element recordings, while
also specific master-to-copy setups are discussed. As the last step in this pure photonic recording
process we explain the bleaching process of these photopolymers and give an insight into digital
holographic recording.

Volume Holographic Optical Elements and Their Applications

vHOEs are thin and lightweight film optics, where the film type influences the optical function.
In contrast to surface gratings vHOEs exhibit an angular and spectral selectivity [1] that can be chosen
by the thickness of the photopolymer film layer. By this selectivity the optics designer can chose the
“On Bragg” condition where the recorded optical function is operative. In contrast no change to the
wave front is observed in all other angles/wavelengths (so called “Off Bragg” conditions). Combined
with the high transparency in the Off Bragg condition vHOEs are an enabler to new key components
like optical combiners for augmented reality glasses [2], large head-up displays [3], and transparent
screens [4]. They can be used as embedded stacked optics for spectrometers [5], sensors and additional
functionalities in existing display technologies as well as new backlight devices for future display
technologies [6,7]. They can be engineered to be ideal for applications where customized filter functions
are needed, e.g., for laser protection [8]. Due to the inherent nature to create complex wave fronts,
vHOEs can also be used as well in the design of luminaires, e.g., for automotive lighting [9]. In large
area applications like new projection display solutions vHOEs can help to improve daylight projection
contrast [10] and can be used to collect light more efficiently for solar cells [11,12]. A classical home
turf for vHOEs is the security market, where their visual effects are a key for recognition without being
obtrusive to the product. From banknotes to ID card to brand protection uses, a wide field of industrial
and consumer applications are in development.
Holographic photopolymers are essential for many future optic designs and complement other
technology developments like light emitting (laser) diodes, spatial light modulators and mobile
computing in compact designs. vHOEs are always linked to full optical system designs, so their
spectral and angular selectivity has to be considered. Advanced vHOEs comprise spatial and/or
spectral multiplexing, which means that several vHOEs can be integrated in one single recording
media film. This ability will drive their adoption into the consumer electronics space. Unique is the
ability to design the diffraction efficiency in wide ranges from sub percent level to close to 100%. In this
review, we discuss the details of this holographic photopolymer technology which is another example
of a beautiful marriage between physics and chemistry.

2. Photopolymers as Recording Media for vHOEs

The need for an easy to use, size scalable recording medium prescribes material properties that
in turn determine the necessary chemistry. A suitable recording medium must not require pre- or
post-processing, has to provide high transparency and low haze after recording and will allow a high
diffractive performance. Additionally, the unrecorded medium has to be manufactured at high quality
standards and the recorded vHOE can be applied in mass production assembly lines. This property
profile can be realized e.g., with a web coated film product.
Several approaches to provide such a material have been reported in the last decades from
industrial as well as from academic laboratories. A major motivation for holographic recording
material developments has been the field of optical data storage and several material concepts have
been described and reviewed elsewhere [13]. While the fundamental physical principles for the
holographic recording of a vHOE or a data page have subtle but significant deviations, the material
design has to focus more densely on the material class that can realize the requirements. For good
reasons, well designed photopolymers are able to fulfill these requirements.
Polymers 2017, 9, 472 3 of 35

2.1. Working Principle

Polymers 2017, 9, 472 of a 2-Chemistry Photopolymer Film 3 of 35

For recording
2.1. Working a stable
Principle of a optical hologram
2-Chemistry function,
Photopolymer Film the recording medium has to have a certain
minimum mechanical modulus to enable a proper grating formation. Liquid recording media tend to
For recording a stable optical hologram function, the recording medium has to have a certain
experience
minimum shrinkage issuesmodulus
mechanical that leadtotoenable
unacceptable
a properoptical
gratingaberrations. In simple
formation. Liquid photopolymer
recording media tendsystems,
a parttoofexperience
the writing chemistry has to be consumed by flood exposure before the
shrinkage issues that lead to unacceptable optical aberrations. In simple photopolymer interference pattern
recording to generate this minimum modulus. By this, a significant part of the dynamic
systems, a part of the writing chemistry has to be consumed by flood exposure before the interference range is already
consumed
pattern and moreover
recording due to the
to generate this pre-exposure
minimum modulus. the whole
By this,recording process
a significant part becomes less reliable.
of the dynamic range
is already
Matrix consumed
formation canandbe moreover
realized by due to the
two pre-exposure
reactive chemistries the whole
that dorecording process
not interfere becomes
with each other
less reliable.
(orthogonal chemistry), so that the writing chemistry has to be unaffected by this matrix formation [14,15].
Matrix
A very importantformation can be realized
advantage of theby2-chemistry
two reactiveapproach
chemistries is that
the do notmiscibility
high interfere with each
of all liquid
other (orthogonal chemistry), so that the writing chemistry has to be unaffected by this matrix
precursors ahead of matrix formation. After matrix formation, no de-mixing of the writing chemistry
formation [14,15].
occurs when carefully designing the components. After hologram recording and curing of the writing
A very important advantage of the 2-chemistry approach is the high miscibility of all liquid
chemistry a stable
precursors interpenetrating
ahead of matrix formation. network
Afterbetween matrix and
matrix formation, writing chemistry
no de-mixing of the writingis formed
chemistry[16].
The raw material of Bayfol ® HX film type photopolymer is composed of two different
occurs when carefully designing the components. After hologram recording and curing of the writing
chemistries,
chemistryasa stable
is shown in Figure network
interpenetrating 1. The between
matrix matrix
precursors, which
and writing are cross-linked
chemistry thermally
is formed [16].
during the Themanufacturing
raw material ofprocess Bayfol® forming
HX film atype polymer networkis and
photopolymer determining
composed of two the physical
different
chemistries,
stability as is shown in Figure
of the photopolymer, 1. The matrix
compose the firstprecursors,
chemistry. whichTheare cross-linked thermally during
imaging components compose
the manufacturing
the second chemistry.process Theseforming
imaging a polymer networkare
components andresponsible
determining the for physical
the hologramstability formation
of the
photopolymer, compose the first chemistry. The imaging components compose
during exposure. They include functionalities for the photo-initiation of the polymerization and the second chemistry.
These imaging components are responsible for the hologram formation during exposure. They
polymerizable groups (monomers). With fully formed matrix, the unexposed photopolymer can be
include functionalities for the photo-initiation of the polymerization and polymerizable groups
stored, shipped and handled easily as long as safe-light conditions are used to prevent accidental
(monomers). With fully formed matrix, the unexposed photopolymer can be stored, shipped and
photo-polymerization.
handled easily as long Byasexposing
safe-light the photopolymer
conditions are used to with a laser
prevent irradiation
accidental interference pattern,
photo-polymerization.
causing
By spatially
exposing distributed
the photopolymer photo-polymerization, the hologram
with a laser irradiation is recorded.
interference pattern, During
causing the exposure
spatially
photo-polymerization is initiated in the bright
distributed photo-polymerization, fringes ofistherecorded.
the hologram laser lightDuring
interference pattern byphoto-
the exposure activating
neighboring imaging
polymerization is components.
initiated in theDuring the exposure
bright fringes the light
of the laser polymerizable
interferencemonomers
pattern by diffuse
activatingwithin
neighboring imaging components. During the exposure the polymerizable
the matrix until they react with an activated monomer or polymer chain. A refractive index modulationmonomers diffuse within
in thethe matrix that
material untilrepresents
they reactthe with an activated
hologram monomer
is caused by theorprocess
polymer of chain. A refractive
polymerization in index
the bright
modulation in the material that represents the hologram is caused by
regions and net diffusion away from the dark regions. For complete hologram formation no furtherthe process of polymerization
in the bright regions and net diffusion away from the dark regions. For complete hologram formation
processing is needed. Photo-curing is only needed to fully bleach the absorbing species to gain optimal
no further processing is needed. Photo-curing is only needed to fully bleach the absorbing species to
transparency and ensuring that the writing chemistry has been fully consumed.
gain optimal transparency and ensuring that the writing chemistry has been fully consumed.

Working Principle of Material

Raw Material
Resin consists of matrix
precursors (●) and
imaging components (●)

Film Production

Unexposed Film
In-situ formation of a
cross-linked matrix

Holographic Exposure

Hologram Holographic exposure to

start photo-polymerization
(recording) process

Figure 1. Working principle of Bayfol® HX film photopolymer [17]: Matrix precursors and imaging
Figure 1. Working principle of Bayfol® HX film photopolymer [17]: Matrix precursors and imaging
components form the raw material resin. The matrix precursors are cross-linked to form a stable
components form the raw material resin. The matrix precursors are cross-linked to form a stable matrix
matrix system hosting the imaging components during film production. This film product is stable
system hosting
until the
exposed to imaging
light. Thecomponents during
film is exposed filmlight
to suitable production.
to initiate This film product is stable
the photo-polymerization that until
exposed to light. The film is exposed to suitable light
forms the hologram structure during the recording process.to initiate the photo-polymerization that forms
the hologram structure during the recording process.
Polymers 2017, 9, 472 4 of 35

By this reaction-diffusion mechanism a pure phase hologram is generated throughout the

thickness d of the photo reactive layer. In the simplest case of recording the interference pattern
of two plane waves a sinusoidal modulation of the refractive index represents this phase grating.
The amplitude of this sinusoidal refractive index modulation is in the following referred to ∆n1 and
represents the performance of the photopolymer under certain recording geometries and recording
conditions [17].
One example of a typical Bayfol® HX film grade for research and development purposes
that was studied in [18] is an acrylate based 2-chemistry photopolymer material and its typical
composition is described in [19]. Such a class of photopolymer can be used for many applications
comprising the fabrication of optical lenses, mirrors, filters, waveguides, diffraction elements and more
complex 3D holographic images. Owing to the high values of refractive index modulation and thus
achievable diffraction efficiency it shows great potential as versatile holographic media. The described
composition in [19] was designed to achieve high diffraction efficiency at ~50 µm film thickness for
reflection recording geometries. This specific photopolymer film has its strengths in applications where
a high spatial frequency and/or high angle and spectral selectivity are required.

2.2. Web-Coating of Bayfol® HX Film

The 2-chemistry approach allows a direct web-coating on a transparent thermoplastic carrier film
(substrate). In a dryer station the thermoset matrix is cured without affecting the writing chemistry.
It is obvious that the coating has to be done in a safe light environment, meaning low brightness and
a selected spectrum to assure that the writing chemistry does not start to polymerize. Photopolymer
layer thicknesses of a few µm up to >50 µm can be coated on transparent substrates of thicknesses
ranging from about 20 µm up to >100 µm. Transparent film substrates are based on polyethylene
terephthalate (PET), polycarbonate (PC), cellulosic triacetate (TAC) and amorphous polyamide (PA).
After curing the photopolymer matrix, the surface of the photopolymer layer is protected with
a removable cover layer. Usually this cover layer is polyethylene (PE) or one of the above-mentioned
substrates as well. After de-lamination of the protective film, the inherent tack of the unrecorded
photopolymer allows excellent optical and mechanical coupling to the recording support surface.
As mentioned, the cover layer can also be chosen from high quality optical films that remain on the
photopolymer surface during hologram recording. In this case, proper optical and mechanical coupling
has to be assured by other means.

3. Modelling the Recording Mechanism of Bayfol® HX Film

Zhao and Mouroulis [20] as well as Noiret et al. [21] for the first time described in 1994 the
holographic recording mechanism in dry photopolymers by the model of photo-polymerization driven
diffusion (PDD). In this PDD model the local monomer consumption rate was set proportional to
the local intensity of the interference field. This means ideal polymerization kinetics is assumed in
the PDD model and moreover the recording mechanism is set to be local in time and space. Despite
these simplifying assumptions the PDD model is able to predict the fact that the achievable index
modulation ∆n1 can drop off significantly if the spatial frequency (SF) of a recorded holographic
grating in a dry photopolymer becomes small. This phenomenon is usually addressed as the low
spatial frequency cutoff.
However many recording materials also experience a high spatial frequency cutoff, which limits
its possible resolution. PDD is not able to predict this high spatial frequency cutoff. For example due
to the finite size of the silver grains in silver halide recording materials there is a limit to the resolution
and by this a high spatial-frequency cutoff. In dry photopolymers no such particles or precipitates exist.
Therefore the reason for an observed high spatial frequency cutoff must be different. By describing the
reaction as a non-local process in time and space the concept of non-local photo-polymerization driven
diffusion (NPDD) was introduced by Sheridan et al. [22]. One channel for non-locality could be, due to
the formation of extended polymer chains or networks, the propagation of the reactive macro-radical
Polymers 2017, 9, 472 5 of 35

in space away from its original position where the chain growth was initiated. A further channel
for non-locality could stem from the primary radicals formed in the photo-initiation system. If those
primary radicals (or very short polymer chains) are too small and of too low reactivity they may diffuse
too fast and start photo-polymerization spatially displaced from the position where the primary radical
was generated. By ascribing a finite resolution length σ, characteristic for the specific photopolymer
system, the NPDD model is able to predict the high spatial frequency cutoff. Within the relevant visible
range of SF (<7500 1/mm corresponding to a Bragg mirror grating for λ = 400 nm in a medium with
refractive index n = 1.5) high resolution photopolymers should have no high spatial-frequency cutoff
to be able to record bright and full-color reflection holograms.
Also implementation of various model improvements with respect to the relation between
monomer consumption rate and local intensity was conducted. For example the relation was changed
from a linear one to a square root behavior [23] by adapting steady state conditions. An extension
to a general power law [24] with an exponent γ in the range of 0.5 to 1 was done in a later stage.
Absorption effects have been included also [25].
Gleeson and Sheridan have given a general review of the work done for free-radical
photo-polymerization in holographic recording in [26]. For limiting recording situations like saturation
recording, short pulse recording and multiplexing general analytical expressions are reported by
Sheridan et al. [27].
In [17] we outlined the saturation recording behavior for the PDD case referring to the scheme
given by Kelly et al. [24]. It is straightforward to extend this to the NPDD case. By assuming that only
the non-reacted monomers are able to move over larger distances or longer times, as done in the above
mentioned reaction-diffusion models, the diffusion equation for the volume fraction of the non-reacted
monomer φMo can be formulated as:

∂φ Mo ∂2 φ Mo
= D· − κ · Pγ · φ Mo · (1 + V · cos(K · x ))γ (1)
∂t ∂x2
For the non-reacted monomer a constant diffusion coefficient D is assumed. The incident recording
power is denoted as P. The various details of the photo-polymerization are summarized in the constant
κ [24]. The wave number of the recorded one dimensional grating is K = 2π/Λ = 2π·SF and related to
the inverse grating spacing Λ. The modulation depth of the interference pattern, also called fringe
visibility is denoted as V.
As we have shown in [17], for saturation recording ∆n1 is only a function of the system parameter
S, characterizing the ratio of the reaction time scale τR over the diffusion time scale τD .

K2 · D 4π2 · SF2 · D
   
∆n1 (S) = ∆n1 = ∆n1 (2)
κ · Pγ κ · Pγ

Equation (2) suggests a scaling behavior. That means a power response curves of ∆n1 taken at
different SF can be combined (shifted) to a master power response curve by appropriate recalibration
of P for each SF. The parameter γ can be identified from the respective (shift) factor α via:
 2/γ
P1 K1
α= = (3)
P2 K2

The excellent match between reaction-diffusion model predictions and experimental results
are depicted in Figure 2 in which the properly calibrated simulations results are compared to the
experimental data [17].
Polymers 2017, 9, 472 6 of 35
Polymers 2017, 9, 472 6 of 35

Figure2.2.Comparison
Figure Comparisonofofthe
theexperimental
experimental(open
(opensymbols)
symbols)and
andsimulated
simulated(filled
(filledsymbols)
symbols)mastered
mastered
powerresponse
power responsecurve
curveafter
afterproper
propercalibration
calibration[17].
[17].

Asshown
As showninin [17]
[17] wewesetset
κexκ=ex1.0
= 1.0
cmcm
2γ/(s·mWγ) for simplicity and identified Dex = 2.5 × 10−10 cm2/s
2γ /(s·mWγ ) for simplicity and identified Dex = 2.5 × 10−10 cm2 /s
in accordance to what was found in [18]. A
in accordance to what was found in [18]. A comparison comparisonofofthe
thesimulated
simulateddosage
dosageresponse
responsecurve
curve∆n
∆n1
1
versusEEavecan
versus canbe
bedone
doneafter
afterhaving
havingnow nowγγand andequivalent
equivalent(D(D/κex/κex) estimated for the photopolymer
ave ex ex ) estimated for the photopolymer
andthe
and theproperly
properlycalibrated
calibratedmodel.
model. The Thefollowing
followingscaling
scalingfor
forthethedosage
dosagebetween
betweensimulation
simulationand
and
experiment can be deduced (note S is kept the same between simulation
experiment can be deduced (note S is kept the same between simulation and experiment): and experiment):

P D
==E ·E ⋅ · ⋅
Eave Paveave D (4)
Eave (4)
P P Dex Dex
Figure
Figure33shows
showsthetheresults.
results.At
Atlarger
largerdosages
dosagesafter
afterthe peak∆n
thepeak ∆n1 1value
valuein
inthe
thesimulation
simulationhas
hasbeen
been
achieved,
achieved,obviously
obviouslythe theexperimental
experimentaldosage
dosageresponse
responsecoincides
coincidesvery
verywell
wellwith
withthe
thepredicted
predicteddosage
dosage
response.
response. AsAsininreal
realsystems
systemsalways
alwaysoxygen
oxygenisispresent.
present.Oxygen
Oxygenquenches
quenchesradicals
radicalsand
andinhibits
inhibitsthe
the
onset
onsetofofphoto-polymerization
photo-polymerizationand andtherefore
thereforethe
theonset
onsetofofgrating
gratingformation.
formation.One Onevery
veryobvious
obviousreason
reason
for
forthe
thediscrepancy
discrepancyatatsmall
smalldosages
dosagesmay maystem
stemfrom
fromthe
thefact
factthat
thatoxygen
oxygenisisnot
notincluded
includedininour
oursimple
simple
model
modelso sofar.
far.
As
Asmentioned
mentionedabove
abovethe
theinclusion
inclusionof ofnon-local
non-localreaction
reactiondiffusion
diffusioneffects
effectscan
canbebeincluded
includedininour
our
simple
simplemodel.
model.Following
FollowingKelly
Kellyetetal.
al.[24]
[24]Equation
Equation(1)(1)has
hastotobe
beextended
extendedto: to:

∂φ ∂2 φ Mo ∂ 2RφMo 1  ( x −(xx−')x2 0)2

 
∞
⋅ φMo· (⋅1(1++ V cos((KK⋅ ·xx'))0 ))⋅ · √ 1 ⋅2exp
∞
= 2D ⋅ − −2 ∞−κ· Pκγ⋅·Pφγ Mo  − − 22·σ2 ⋅ dx·' dx0
γ γ
∂φ Mo
= D Mo
· ∂x V ⋅·cos · exp (5)
(5)
∂t ∂x  2⋅σ 
σ·2σ
∂t −∞
2π ⋅2π  
ForVVequal
For equaltoto11and
andininthe
thecase
caseofofsaturation
saturation recording
recordingwewe then find
then find∆n∆n
1 =1 ∆n 1 (S;
= ∆n K2K·2σ⋅σ2 ).
1(S; 2).The
Thehigh
high
Now∆n
2 2
spatial
spatialfrequency
frequencycutoff
cutoffisisdescribed
describedherehereasasKK·σ⋅σ. .Now
2 2 ∆n11 starts
starts to
to drop
dropdowndowndue dueto toinsufficient
insufficient
resolution
resolutionofofthe photopolymerififΛΛ==1/SF
thephotopolymer 1/SF is
is reduced
reduced and andapproaches
approachesthe thesame
sameorderorderofofmagnitude
magnitude
as σ. We conducted respective model calculations for P = 8 mW/cm 2 2with varying non-locality σ2 2
as σ. We conducted respective model calculations forave Pave = 8 mW/cm with varying non-locality σ
using
usingthe
theidentified
identifiedsystem
systemparameters
parametersγ,γ,(D (Dexex/κ ex)) and
/κex and the
the identified
identified calibration
calibration of ∆n ∆n11. .
Polymers 2017, 9, 472 7 of 35
Polymers 2017, 9, 472 7 of 35
Polymers 2017, 9, 472 7 of 35

Figure 3.
Figure 3. Comparison
Comparison of
of dosage
dosage response
response curves
curves for
for simulated
simulated data
data and
and experimental
experimental data
data for
for similar
similar
recording power
recording
recording power densities
power densitiesPP
densities Pave
ave [17]. Note for the same Pave simulation and
simulationand
avesimulation
ave [17]. Note for the same Pave
experiment
andexperiment should
experiment should coincide.
should coincide.

The following
following Figure 4 shows the results with the
the two
two experimental
experimental ∆n ∆n1 values added which
The following Figure 4 shows the results with the two experimental ∆n11 values added which
were deduced from plane wave plane wave
were deduced from plane wave plane wave recordings
wave recordings [17]
recordings [17] at
[17] at P ave = 8 mW/cm22.2
at PPave
ave == 88 mW/cm
mW/cm. .

Figure 4. Calculated ∆n ∆n1 versus Λ = 1/SF

1/SF for
for varying
varying non-local parameters
parameters σ2 andand PPave
ave = 8 mW/cm222.
Figure 4. Calculated ∆n11 versus Λ == 1/SF for non-local parameters σσ22 and Pave == 8 mW/cm
mW/cm ..
The filled symbols indicate the respective
The filled symbols indicate the respective experimental
respective experimental values
experimental values measured
values measured at
measuredat
atSFSF = 1429
SF== 1429 L/mm
1429 L/mm (dashed
L/mm (dashed
line) and SF = 4525
line) and SF == 4525 L/mm
4525 L/mm (dotted
L/mm(dotted line)
(dottedline) [17].
line)[17].
[17].

In Figure
In Figure 44 the
the dashed
dashed and and the dotted
dotted lines
lines refer
refer to the
the respective
respective SF SF = 1429
1429 1/mm
1/mm and and 4525
4525
In Figure 4 the dashed and thethedotted lines refer to thetorespective SF = 1429=1/mm and 4525 1/mm.
1/mm. We can
1/mm. can conclude
conclude from
from Figure 44 that that σ
σ2 should
2 should be be in
in the
the range
range of of <100
<100 nmnm2 . For
2 For aa variety
variety of
We canWe conclude from FigureFigure 4 that σ2 should be in the range of <100 nm2 . .For a variety of of
photopolymers
photopolymers σ 2 values
σ2 values are estimated in [22] and the smallest values cited are larger than 2000 nm
nm22.
values are
are estimated
estimated®in in [22]
[22] and
and the
the smallest
smallest values
values cited
cited are
are larger
larger than
than 2000
2000 nm2..
2
photopolymers σ
These results underline
These underline thatthat for
for Bayfol HX we do
Bayfol®® HX do not have
have to expect
expect a significant
significant drop
drop of
of ∆n
∆n1 due to
to
These results
results underline that for Bayfol HX wewe do notnot have to to expect aa significant drop of ∆n11 due
due to
aaa high
high spatial frequency
high spatial
frequency cutoff at
spatial frequency cutoff
at all SF
cutoff at all
SF relevant
relevant for
all SF relevant
for visible
visible light
for visible
light (e.g.,
(e.g., for
light (e.g.,
for RGB
RGB reflection
for RGB
reflection holograms
holograms
reflection holograms
SFSF ~ 4000–7500 1/mm).
SF ~~ 4000–7500
4000–7500 1/mm).
1/mm).
The absence
The absence of of a high
high spatial
spatial frequency
frequency cutoff
cutoff was further
further confirmed
confirmed in in aa study
study inin which
which a
The absence of a ahigh spatial frequency cutoff waswasfurther confirmed in a study in which a Bayfola®
Bayfol® HX
Bayfol
® HX film
film typetype photopolymer
photopolymer was was compared
compared to to the
the widely
widely studied
studied acrylamide
acrylamide based
HX film type photopolymer was compared to the widely studied acrylamide based photopolymerbased with
photopolymer with respect to their experimentally measured
photopolymer with respect to their experimentally measured and simulated spatial and simulated spatial frequency
frequency
response [18].
response [18]. The
The results
results are
are depicted
depicted inin Figure
Figure 5.
5.
Polymers 2017, 9, 472 8 of 35

respect to their experimentally measured and simulated spatial frequency response [18]. The results
are depicted in Figure
Polymers 2017, 9, 472 5. 8 of 35

FigureFigure 5. Experimentalvalues
5. Experimental values of
of ∆n
∆n11(symbols)
(symbols)versus grating
versus spacing
grating Λ (or spatial
spacing Λ (or frequency SF ~
spatial frequency
1/Λ) for experimental Bayfol ® HX film type photopolymers in comparison to an acrylamid based
®
SF ~ 1/Λ) for experimental Bayfol HX film type photopolymers in comparison to an acrylamid
photopolymer system [18]. Extracted values for σ2 are 85 nm2 for the experimental Bayfol® HX film
based photopolymer system [18]. Extracted values for σ2 are 85 nm2 for the experimental Bayfol® HX
type photopolymer and 3840 nm2 for the acrylamid based photopolymer.
film type photopolymer and 3840 nm2 for the acrylamid based photopolymer.
Again for the experimental Bayfol® HX film type photopolymer a value for σ2 of 85 nm2 could
Again for thewhereas
experimental ® 2HX film type photopolymer a value for σ2 of 85 nm2 could be
be extracted a σ2 of Bayfol
3840 nm was found for the acrylamid based photopolymer. To relate
thesewhereas
extracted numbersatoσa of2 grating
3840spacing2
nm was Λ of a retro-reflecting
found Bragg mirror
for the acrylamid based for λ = 400 nm in To
photopolymer. a medium
relate these
with to
numbers a arefractive
grating index
spacing Λ 1.5
n of of awhich is 133 nm, Bragg
retro-reflecting the non-locality
mirror forσ λin= a400
Bayfol
nm®inHX film typewith
a medium
photopolymer
a refractive index n isof in
1.5the rangeisof
which 1339–10
nm,nmthewhich is still much
non-locality σ in asmaller
Bayfolthan
® HX 133 nm.
film In an
type acrylamid is
photopolymer
based photopolymer this smearing σ of the localization of the interference pattern is in the range of
in the range of 9–10 nm which is still much smaller than 133 nm. In an acrylamid based photopolymer
62 nm which almost half of the respective grating spacing. This large σ explains why in such a type
this smearing σ of the localization of the interference pattern is in the range of 62 nm which almost
of photopolymer the recording of a high spatial frequency grating shows a very low index
half ofmodulation
the respective
∆n1. grating spacing. This large σ explains why in such a type of photopolymer the
recording of a high spatial frequency grating shows a very low index modulation ∆n1 .
Recording at the Low Spatial Frequency Cutoff
Recording at the Low Spatial Frequency Cutoff
At a first glance the drop of ∆n1 with decreasing SF or increasing P, due to the low spatial
At a firstcut
frequency glance
off maythe drop
appear of ∆n1 with
as drawback decreasing SF
for photopolymers. or increasing
However P, was
this ∆n1 drop dueusedto very
the low
spatial frequencytocut
successfully off high
record mayquality
appearvHOEas drawback
gratings infor photopolymers.
Bayfol However this ∆n1 drop
® HX film type photopolymers. Those

transmission
was used gratings are used
very successfully to for spectral
record analysis
high in astronomical
quality vHOE gratingstelescopes
in [28]. For®transmission
Bayfol HX film type
photopolymers. Those transmission gratings are used for spectral analysis of
gratings over-modulation can cause difficulties in managing the optical performance ina astronomical
vHOE in
practical recording. This is due to the fact that the product ∆n1⋅d and by this the diffraction efficiency
telescopes [28]. For transmission gratings over-modulation can cause difficulties in managing the
that is proportional to sin2 (Constant⋅∆n1⋅d) may vary strongly with the recording dosage E. Constant
optical performance of a vHOE in practical recording. This is due to the fact that the product ∆n1 ·d
represents the recording geometry. Therefore the dosage margin to hit the maximum diffraction
and by this the diffraction efficiency that is proportional to sin2 (Constant·∆n1 ·d) may
efficiency can become very tight. Due to the variation of ∆n1 via P the specific material
vary strongly
can be tuned
with the recording
easily dosage
to maximum E. Constant
diffraction represents
efficiency the recording
for transmission geometry.
gratings Therefore
at ideal the dosage
fringe visibility margin
V for
to hit saturated
the maximum
dosagediffraction efficiency
E. This results candosage
in largest become very for
margins tight.
theDue to theprocess
recording variation
andofV ∆n
close to 1P the
1 via
specific material
reveals can
lowest be in
noise tuned easily tograting.
the recorded maximum diffraction efficiency for transmission gratings at
ideal fringe visibility V for saturated dosage E. This results in largest dosage margins for the recording
process4. Optimization
and V close to of 1the Photo-Initiator
reveals System
lowest noise for recorded
in the the Visible Spectrum
grating.
For a photopolymer material used in holographic recording its spectral sensitivity has to be
4. Optimization of wavelength
aligned to the the Photo-Initiator
λ of laserSystem for the
light used Visible Spectrum
for recording to enable a high efficiency of the
conversion
For of the photon’s
a photopolymer energy used
material resulting in an overall economically
in holographic feasible
recording its recording
spectral step. Due
sensitivity has to
to the limited power of highly coherent lasers, holographic recording costs are heavily influenced by
be aligned to the wavelength λ of laser light used for recording to enable a high efficiency of
the overall efficiency of the photonic steps. This overall efficiency is determined by the molecular
the conversion of the photon’s energy resulting in an overall economically feasible recording
absorption coefficient of the sensitizer dyes at the recording wavelength λ, the quantum yield with
step. Due to the limited power of highly coherent lasers, holographic recording costs are
Polymers 2017, 9, 472 9 of 35

heavily influenced by the overall efficiency of the photonic steps. This overall efficiency is
determined by the molecular absorption coefficient of the sensitizer dyes at the recording wavelength
λ, the quantum yield with which the reactive species are formed and react, the reactivity of
generated photo products, and the possibility of the bleaching of the sensitizer dyes during the
Polymers 2017, 9, 472
photo-polymerization [13,29–32]. The photo-polymerization efficiency thus depends on 9the of 35nature

of photo-initiators.
Polymers 2017, 9, 472 Photo-initiating systems for the visible spectrum that were developed9 in recent
of 35
which the reactive species are formed and react, the reactivity of generated photo products, and the
years include ketocoumarines, functional dyes, thioxanthones, bis-acylphosphine
possibility of the bleaching of the sensitizer dyes during the photo-polymerization [13,29–32]. The oxides, pyrylium
which the reactive
andphoto-polymerization
thiopyrylium species
salts, are formed
(cationic) dyes andinreact, the reactivity
combination of generated
with a boratephotosalt, products, and the
dyes/bis-imidazole
efficiency thus depends on the nature of photo-initiators. Photo-initiating
possibility of the bleaching of the sensitizer dyes during the photo-polymerization [13,29–32]. The
derivatives/thiols,
systems for the visiblephotosensitizer/chlorotriazine/additives,
spectrum that were developed in recent etc.years
[33,34].
includeThisketocoumarines,
diverse offering of
photo-polymerization efficiency thus depends on the nature of photo-initiators. Photo-initiating
photo-active
functional chemicals for phase bis-acylphosphine
dyes, thioxanthones, hologram recording poses
oxides, the taskand
pyrylium to thiopyrylium
select the best in class.
salts,
systems for the visible spectrum that were developed in recent years include ketocoumarines,
(cationic)
Furthermore, dyes in combination with a borate salt, dyes/bis-imidazole derivatives/thiols,
functionala crucial necessity for the
dyes, thioxanthones, photo active substances
bis-acylphosphine is to do their
oxides, pyrylium and job traceless, i.e.,
thiopyrylium for any
salts,
photosensitizer/chlorotriazine/additives,
optical etc. [33,34]. coloration,
This diverse stain
offering of photo-active chemicals
(cationic) dyes in combination with a borate salt, dyes/bis-imidazole derivatives/thiols, by
application the visible range, no remaining or even hue will be accepted
for phase hologram recording poses the task to select the best in class. Furthermore, a crucial necessity
the end-user.
photosensitizer/chlorotriazine/additives, etc. [33,34]. This diverse offering of photo-active chemicals
for the photo active substances is to do their job traceless, i.e., for any optical application the visible
From a mechanistic
for phase approach,
hologram recording thethe
poses photo-initiation
task to select theofbest
a radical polymerization
in class. Furthermore, a can occur
crucial by a direct
necessity
range, no remaining coloration, stain or even hue will be accepted by the end-user.
for the photo
or a sensitized active substances
decomposition of is
thetophoto-initiator.
do their job traceless, i.e., for any
According optical
to this, application thesystems
photosensitive visible can
From a mechanistic approach, the photo-initiation of a radical polymerization can occur by a
range, no remaining
be classified coloration, stain or even hue will be accepted
with by the end-user.
direct or aas one, two
sensitized or even three-component
decomposition systems,
of the photo-initiator. According the one-component
to this, photosensitivesystem
systemshaving
all required From a mechanistic
functionalities approach,
being the
encoded photo-initiation
in one single of a radical
molecule polymerization
and the three can occur bysystem
component a
can be classified as one, two or even three-component systems, with the one-component system
direct or a sensitized decomposition of the photo-initiator. According to this, photosensitive systems
using threeallindividual
having molecules tobeing
required functionalities fulfillencoded
the task. From
in one thismolecule
single simplistic andpoint of view,
the three the material
component
can be classified as one, two or even three-component systems, with the one-component system
development
system using hasthree
either to prepare
individual the perfect
molecules molecule
to fulfill or toFrom
the task. select a smart
this combination
simplistic of two
point of view, the[35] or
having all required functionalities being encoded in one single molecule and the three component
threematerial development
components [36]. has either
Since there toisprepare
good the perfect
evidence formolecule
all these orconcepts,
to select a one
smart
system using three individual molecules to fulfill the task. From this simplistic point of view, the one-
combination
example for of
each
two [35] or
and two-component three components
system [36]. Since there is good evidence for all these concepts, one example
material development hasiseither
discussed first.the perfect molecule or to select a smart combination of
to prepare
for each one- and two-component system is discussed first.
two [35] or three components [36]. Since there is good evidence for all these concepts, one example
4.1. One-Component
for each one- andSystem
two-component system is discussed first.
4.1. One-Component System
In this system, irradiation of the photo initiator (PI) provides the reactive species for the initiation
4.1. In
One-Component System of the photo initiator (PI) provides the reactive species for the initiation
this system, irradiation
step. In more detail the PI absorbs the light energy and transforms intra molecularly into a reactive
step. In more detail the PI absorbs •the light energy and transforms intra molecularly into a reactive
In this
ansystem, irradiation ofRthe photo initiator (PI) provides the reactive species for the initiation
species, here
species, here initiating
an initiatingradical that
radical R that attacks
attacks a amonomer
monomer MM andand finally
finally converts
converts into
into the the (photo)
(photo)
step.
polymer In more
(Scheme detail
1): the PI absorbs the light energy and transforms intra molecularly into a reactive
polymer (Scheme 1):
species, here an initiating radical R that attacks a monomer M and finally converts into the (photo)
polymer (Scheme 1): hν ∗ • M
PI ⎯⎯→PI ⎯ ⎯→R ⎯⎯→Polymer
∗
Scheme 1.PI
hν
⎯⎯→
Reaction ofPI ⎯→R• ⎯⎯
⎯
a one-component
M
→Polymer
photo-initiator system.
Scheme 1. Reaction of a one-component photo-initiator system.
Scheme 1. Reaction of a one-component photo-initiator system.
Titanocene derivatives represent one of the few examples of photo-initiators for the visible
Titanocene
spectral rangederivatives represent
(VIS) that are directlyone of the few
photolyzed examples
upon of photo-initiators
light exposure. More examples for the
can visible
be foundspectral
Titanocene derivatives represent one of the few examples of photo-initiators for the visible
range (VIS)The
in [36]. that
useare directly photolyzed
of titanocene upon
complexes (e.g., light exposure.
CGI-784 from BASF More examples can
AG, Ludwigshafen, be found
Germany, seein [36].
spectral range (VIS) that are directly photolyzed upon light exposure. More examples can be found
The Scheme
use of 2) in experiments
titanocene where
complexes several
(e.g., high refractive
CGI-784 from BASFindex
AG,organic monomersGermany,
Ludwigshafen, were incorporated
in [36]. The use of titanocene complexes (e.g., CGI-784 from BASF AG, Ludwigshafen, Germany, see see Scheme 2)
into acrylate
in experiments oligomer-based
where formulations,
severalwhere
high refractive irradiated
index at 546
organic nm was
monomersdescribed in [37].
were incorporated into acrylate
Scheme 2) in experiments several high refractive index organic monomers were incorporated
oligomer-based
into acrylateformulations,
oligomer-basedirradiated at 546
formulations, nm was
irradiated at described
546 nm wasindescribed
[37]. in [37].

Scheme 2. Structure formula of the titanocene complex CGI-784.

Scheme 2. Structure formula of the titanocene complex CGI-784.

The presence or absence2.ofStructure
Scheme oxygen (in N2) plays
formula a vital
of the role in the
titanocene initiation
complex process as was shown
CGI-784.
by studies with different concentrations of titanocene and at different incident light intensity. In
The presence or absence of oxygen (in N2) plays a vital role in the initiation process as was shown
aerobic and anaerobic atmospheres it was observed that the polymerization reaction commenced
by studies with different concentrations of titanocene and at different incident light intensity. In
rapidly. After reacting for a few seconds, this process however suffered abrupt deactivation, being
aerobic and anaerobic atmospheres it was observed that the polymerization reaction commenced
more pronounced at low titanocene concentration [38]. Additionally, photo bleaching of the
rapidly. After reacting for a few seconds, this process however suffered abrupt deactivation, being
more pronounced at low titanocene concentration [38]. Additionally, photo bleaching of the
Polymers 2017, 9, 472 10 of 35

The presence or absence of oxygen (in N2 ) plays a vital role in the initiation process as was shown
by studies with different concentrations of titanocene and at different incident light intensity. In aerobic
and anaerobic
Polymers atmospheres it was observed that the polymerization reaction commenced rapidly.10After
2017, 9, 472 of 35
reacting for a few seconds, this process however suffered abrupt deactivation, being more pronounced
titanocene
at requires
low titanocene rather harsh
concentration conditions
[38]. like high
Additionally, photoUV dosagesof which
bleaching lead to requires
the titanocene some residual
rather
coloration
harsh in the like
conditions recorded
high UVmaterial.
dosagesThe addition
which lead toofsome
well-known
residual additives
colorationinto
in thetherecorded
photo initiating
material.
system
The (i.e., diphenyliodonium
addition salt, N-methyldiethanolamine,
of well-known additives into the photo initiatingetc.) raises
system thediphenyliodonium
(i.e., efficiency of the photo
salt,
initiation [39]. These additives utilize the photo chemically generated radicals that may
N-methyldiethanolamine, etc.) raises the efficiency of the photo initiation [39]. These additives utilize inhibit the
polymerization
the photo chemicallyprocess to form new
generated radicals
radicals that active in polymerization.
may inhibit As theseprocess
the polymerization additives now play
to form new
a vital role
radicals in the
active photo initiating system
in polymerization. As these they transform
additives nowitplay
virtually into
a vital a in
role two thecomponent photo
photo initiating
initiator.
system they transform it virtually into a two component photo initiator.

4.2. Two-Component System

System Dye/Organo-Borate
Dye/Organo-Borate Salt
Salt
Two-component
Two-componentphoto photoinitiating
initiating systems
systems areare
the the
commonly
commonly usedused
material combination
material that allow
combination that
for
allowan for
efficient photo-polymerization
an efficient photo-polymerization in the VIS. They
in the contain
VIS. one component
They contain to absorbtothe
one component light the
absorb for
example a sensitizer
light for example dye and one
a sensitizer dye component that interact
and one component thatwith the sensitizer
interact dye to form
with the sensitizer dyethe
toinitiating
form the
species.
initiatingInspecies.
generalIn they followthey
general a bimolecular and a stepwise
follow a bimolecular and process
a stepwiseconsisting
processofconsisting
a first stepofofalight
first
absorption
step of lightby the photo-initiator
absorption or dye (i.e., thioxanthones,
by the photo-initiator ketocoumarines,
or dye (i.e., thioxanthones, etc.) and a second
ketocoumarines, etc.) andstepa
by atomstep
second or electron
by atom transfer through
or electron a co-initiator
transfer through(i.e., alcohols or (i.e.,
a co-initiator thiolsalcohols
serve asor hydrogen donors
thiols serve as
and/or
hydrogen amines
donors serve as electron
and/or aminesdonors)
serve asthat subsequently
electron donors) generates the initiating
that subsequently species.
generates theEspecially
initiating
for electron
species. transfersfor
Especially photo-initiators
electron transfers an additional way ofan
photo-initiators deactivation
additionalhas way toofbedeactivation
considered—the has tobackbe
electron transfer.
considered—the back electron transfer.
Combinations
Combinations of ofselected
selecteddyesdyeswithwithtriarylalkyl-borate
triarylalkyl-borate salts have
salts havebeen
beenreported
reported as co-initiators
as co-initiatorsfor
Norrish
for Norrishtype type
II photo-initiating
II photo-initiatingsystems [40]. The
systems initiation
[40]. mechanism
The initiation of the radical
mechanism of thechain formation
radical chain
involves
formationa involves
photo-induced electron transfer
a photo-induced electronfrom the from
transfer borate thesalt to the
borate saltsinglet
to the excited state (1 Dye)
singlet excited state
or triplet 3
state (state
Dye)( Dye)
of theofdye [41].[41].
Subsequently, thethe
boranyl radical R3RR’B •
( Dye)
1 or triplet 3 the dye Subsequently, boranyl radical 3R’B is
• isformed,
formed, which
may undergo a carbon-boron
carbon-boron bond bond dissociation
dissociation generating
generating an an alkyl
alkyl radical
radical (Scheme
(Scheme 3). 3). The rate
constant for the mentioned electron transfer reaction correlates with the Gibbs free energy change
(∆G
(ΔGet).). Furthermore,
Furthermore, the the rate
rate of
of the
the C–BC–B cleavage
cleavage depends
depends on the thermodynamic stability of the
primary radical formed.
formed. For example the lifetime of the triphenyl-butyl boranyl radical is reported to
be the range of 250 fs [42], preventing any non-productive back electron transfer reaction. A potential
dependency of the polymerization efficiency on the structure of the borate salt is still contradictory
discussed, since studies exist with a correlation while other studies reveal that a specific behavior is
not fully understood [43,44].

Scheme 3.
Scheme 3. Reaction
Reaction of
of aa two-component photo-initiator system.
two-component photo-initiator system.

A good way to investigate and characterize the photo chemistry of these type (II) photo initiators
A good way to investigate and characterize the photo chemistry of these type (II) photo initiators
is to study the decay of an acrylic reactive thinner that readily dissolves the photo initiator by online
is to study the decay of an acrylic reactive thinner that readily dissolves the photo initiator by online
IR spectroscopy. The obtained kinetic data allow us to deduce potential reaction mechanism, relative
IR spectroscopy. The obtained kinetic data allow us to deduce potential reaction mechanism, relative
reactivities and chemical stabilities. Rates of conversion Rp and final conversion C obtained from the
reactivities and chemical stabilities. Rates of conversion Rp and final conversion C obtained from the
photo-polymerization experiments of an acrylate resin under laser irradiation at 532 nm are collected
photo-polymerization experiments of an acrylate resin under laser irradiation at 532 nm are collected
in Table 1, using the borate salts given in Scheme 4.
in Table 1, using the borate salts given in Scheme 4.
Polymers 2017, 9,
Polymers 4729, 472
2017, 11 of 35
11 of 35

Polymers 2017,
Table 1. 9, 472
Oxidation potentials Eox, rates of polymerization Rp and final conversions C of the borate salts11 of 35
1. Oxidation
Tableused potentials Eox , rates
in the photo-polymerizing resinof polymerization
with Rp and
Rose Bengal (RB) final conversions
or Safranine C of[35].
O (SF) as dye the borate salts
Table
used in the 1. Oxidation potentials resin
photo-polymerizing Eox, rates
withof polymerization
Rose Bengal Rp and
(RB) or final conversions
Safranine O (SF)C as
of the
dye borate
[35]. salts
Borate
used in the E ox (V/SCE)
photo-polymerizingRB Rp (M
resin with∙sRose
−1 −1 ) Bengal C (%)or Safranine
(RB) SF Rp O (M(SF)
−1 −1 ∙s as
) dye [35].
C (%)
1 0.23 1.20 80 0.50 61
−1 −1 −1 ·−1s−1 )
Borate
Borate
2 EoxEox (V/SCE)
(V/SCE)
0.38 RBRRp0.76
RB (M−1∙s·s−1) )
p(M CC77
(%)
(%) SF R
SF (M−1
Rpp1.2
(M ∙s ) C67 C (%)
(%)
1 31 0.23
0.49
0.23 1.20
0.16
1.20 80
72
80 0.50
1.8
0.50 61
74 61
2 42 0.38
0.38
0.72 0.76
0.76
0.046 77
77
63 1.2
1.51.2 67 67
72
3 53 0.49
0.49
0.74 0.16
0.16
0.020 72
72
49 1.8
1.8
0.94 74 74
70
4 64 0.72
0.72
0.81 0.046
0.046
0.0038 63
63
15 1.5
1.5
1.3 72 72
71
55 0.74
0.74 0.020
0.020 49
49 0.94
0.94 70 70
66 0.81
0.81 0.0038
0.0038 15
15 Ph 1.3
1.3 71 71
Bu Bu
Bu B Bu Bu B Ph Ph B
Bu Bu Ph
Ph
Bu Bu tBu 3B
1B Bu 2B Ph Ph
Bu Bu Ph
Bu Bu
1 2 tBu 3

B Bu B Bu
But B Bu

B Bu B Bu
But B Bu
4 5
But 6
4 5
But(counter
Scheme 4. Borate anions used in this study 6 4 or +NBu4).
ions are +NMe

The high Scheme

reactivity 4. Borate anions
of anions
borate used
salts in
inas this study(counter
(counter
co-initiators ions are
is demonstrated+ NMe
are+NMe or
4 due + 4their
+NBu ). 4 ). high final
Scheme 4. Borate used this study ions 4 or toNBu
conversions for most of the systems used. For the Rose Bengal/organo-borate systems Rp values
The continuously
decrease high reactivity withof increasing
borate salts Eoxas(Figure
co-initiators
6). Theismaximum
demonstrated
Rp was due to theirforhigh
detected final
1 which
The high reactivity
conversions for most of the
of borate saltsused.
systems as co-initiators
For the Rose is demonstrated duesystems
Bengal/organo-borate to their R high final
p values
exhibits the least Eox. An about 300 times reduced Rp value was determined for 6 which exhibits a
conversions
decrease for most of the systems used. For the Rose Bengal/organo-borate systems Rp values
higher Eox continuously
than that of 1.with increasing Eox (Figure 6). The maximum Rp was detected for 1 which
decrease continuously
exhibits the least E with
ox. An increasing
about Eox (Figure
300 times
This is consistent with the assumption that the6).
reduced The maximum
Refficiency
p value was Rp was
determined
of the for 6detected for 1 which
which exhibits
photo-polymerization reactiona
exhibits the
higher least
Eox thanE .
that An
of 1.about 300 times
depends mainlyoxon the primary electron transfer process.reduced R p value was determined for 6 which
In the case of Safranine O/organo-borate exhibits
salts,EThis
a higher is consistent
Rp than
ox does of 1.withwith
thatcorrelate
not the Eassumption that the efficiency of the photo-polymerization reaction
ox of the borates (Figure 6) and Rp varies only by a factor of 4.
depends mainly on the primary electron transfer process. In the case of Safranine O/organo-borate
salts, Rp does not correlate with Eox of the borates (Figure 6) and Rp varies only by a factor of 4.

Figure 6. Rate of polymerization Rp for Rose Bengal (RB) (filled symbols) and Safranine O (SF) (open
Figure 6. Rate of polymerization Rp for Rose Bengal (RB) (filled symbols) and Safranine O (SF) (open
symbols) versus the oxidation potential Eox of the borate salts [35].
symbols) versus
Figure theofoxidation
6. Rate polymerization Rp forERose
potential ox of Bengal
the borate
(RB) salts
(filled[35].
symbols) and Safranine O (SF) (open
symbols)
The versusisthe
reactivity oxidation
very high forpotential
all the E ox of the borate salts
organo-borates, [35].
quite close to the highest Rp found for Rose
Bengal/organo-borate
This is consistent with complexes. Thus, Rose
the assumption thatBengal and Safranine
the efficiency of the O significantly vary in their
photo-polymerization reaction
The reactivity
reactivity as aonpart is very
a typehigh for all the organo-borates, quite close to the highest Rp found for Rose
depends mainly theofprimary II photo-initiating
electron transfer system.
process. In the case of Safranine O/organo-borate
Bengal/organo-borate complexes. Thus, Rose Bengal and Safranine O significantly vary in their
salts, R p does not
reactivity
correlate with Eox of the borates
as a part of a type II photo-initiating
(Figure 6) and Rp varies only by a factor of 4.
system.
The reactivity is very high for all the organo-borates, quite close to the highest Rp found for
Rose Bengal/organo-borate complexes. Thus, Rose Bengal and Safranine O significantly vary in their
reactivity as a part of a type II photo-initiating system.
Polymers 2017, 9, 472 12 of 35

4.3. Excited State Process

The Gibbs free energy ∆Get accompanying the electron transfer reaction appearing within the
dye/organo-borate combinations can be estimated using the Rehm-Weller equation [45]

∆Get
S = E −E
ox red − ES
(6)
∆Get = Eox − Ered − ET
T

With ∆Get S and ∆Get T defined as Gibbs free energy for the electron transfer to the excited
singlet state and excited triplet state, respectively. Ered is defined as reduction potential of the
dye, ES and ET for the singlet or triplet energy of the dye. The Rehm-Weller equation neglects
the coulomb term. The corresponding values for all possible dye/organo-borate combinations are
thermodynamically favorable. Therefore it deserves a more detailed mechanistic study to understand
the excited state’s reactivity.
Fluorescence spectroscopy was used to study the singlet excited states of the dyes and laser
flash photolysis (LFP) for their triplet excited states. The corresponding excited state quenching rate
constants (kq S or kq T ) are accessed from the Stern-Volmer analysis of the decay curves

1 1
= + kS,T
q · [ Q] (7)
τ τ0

τ and τ0 are the lifetimes of the excited states with or without the quencher Q—the organo-borate salt
in this case.
The rate constant kq S of the reaction of the singlet excited states 1 RB and 1 SF with the
organo-borate salts was detected using fluorescence spectroscopy with excitation at 550 nm for RB
and 460 nm for SF (Table 2). 1 SF leads to more or less diffusion controlled reactivity in acetonitrile as
solvent, which is in agreement with the rather negative values for ∆Get S . In the opposite case of 1 RB,
although the ∆Get S values are rather low, much more reduced kq S values (below the diffusion limit)
were observed.
Laser flash photolysis studies were done at 532 nm in acetonitrile for both RB and SF. The
lifetime of 3 SF decreases when adding the organo-borate salt which was monitored at 830 nm. The
associated quenching rate constants kq T are diffusion controlled for the most negative values of ∆Get T
(i.e., for 4) and are diminished by a factor of 20 for 6 (Table 2). The ground state photo-bleaching of
Safranine O at relatively long time scales, which was monitored at 510 nm, displays a higher population
with increasing concentration of organo-borate salt. This can be interpreted as a consequence of the
formation of the semi-reduced form of Safranine O. The latter transient species was detected at 430 nm
and displays an increase with increasing concentration of borate salt. This is the result of the reactivity
of both the singlet and triplet excited states of Safranine O with the organo-borate salts.

Table 2. Quenching rate constants kq S,T (M−1 ·s− 1 ) for the excited states of the dye in acetonitrile and
the corresponding Gibbs free energies ∆Get (eV) [35].

Borate kq S /∆Get S 1 RB kq T /∆Get T 3 RB kq S /∆Get S 1 SF kq T /∆Get T 3 SF

1 1.8 × 109 /−0.94 2.2 × 106 /−0.57 1.8 × 1010 /−1.56 1.8 × 1010 /−0.98
2 1.6 × 109 /−0.79 1.5 × 106 /−0.42 1.8 × 1010 /−1.41 1.3 × 1010 /−0.83
3 a /−0.68 a /−0.31 1.2 × 1010 /−1.30 8.2 × 109 /−0.72
4 8.0 × 108 /−0.45 6.3 × 105 /−0.08 2.0 × 1010 /−1.07 6.8 × 109 /−0.49
5 7.3 × 108 /−0.43 7.7 × 105 /−0.06 1.4 × 1010 /−1.05 3.8 × 109 /−0.47
6 4.8 × 108 /−0.36 4.3 × 105 /−0.01 1.6 × 1010 /−0.98 9.4 × 108 /−0.40
a Not measurable due to insolubility at high concentrations of 3.
Polymers 2017, 9, 472 13 of 35

The triplet state of Rose Bengal can be monitored by LFP at 610 nm in acetonitrile. The observed
data for2017,
Polymers the 9,different
472 triplet state quenching rate constants are collected in Table 2. The kq T values 13 of 35
behave inversely proportional with ∆Get T . More interestingly, even with such a diminished value as
6 6M−1−1·s−
− 0.57 eV for
−0.57 for ∆G∆GetetTT a low
low quenching
quenchingrate rateconstant
constantofof2.2 2.2× ×1010 M ·s−11 isis observed,
observed, i.e., aa ~300 ~300 times
times
reduced reactivity than for Safranine O at similar ∆G TT (see 33 SF/4, kT T = 6.8 × 9109 −1M− 1 ·s−1 ).
reduced reactivity than for Safranine O at similar ∆Getet SF/4, kq q = 6.8 × 10 M ·s ). −1

Under
Under excitation
excitation by by light
light Rose
Rose Bengal
Bengal therefore
therefore reacts
reacts with
with the
theorgano-borate
organo-borate salts salts from
fromboth
both
the
thesinglet
singletand andthe
thetriplet
tripletstate.
state. This
Thisresults
resultsin inaaground
groundstatestatephoto-bleaching
photo-bleachingof ofRose
RoseBengal
Bengalwhichwhich
can
canbebedetected
detectedat at550550nm.
nm.Concomitantly,
Concomitantly,the theRose
RoseBengal
Bengalsemi-reduced
semi-reducedform, form,which
whichisisgenerated
generated
during
duringthisthisreaction,
reaction, was detected
was at 430
detected at nm
430innm acetonitrile, provingproving
in acetonitrile, the electron transfer. The
the electron obtained
transfer. The
quenching rate constants
obtained quenching of the singlet
rate constants of theand the triplet
singlet excited
and the states
triplet of Rose
excited statesBengal
of Rose increase
Bengalwith the
increase
decreasing of the oxidation
with the decreasing potential of
of the oxidation organo-borate
potential salts. Thesesalts.
of organo-borate results are results
These in line with
are inthelinemeasured
with the
photo-initiating efficiency ofefficiency
measured photo-initiating the respective
of thereaction
respectivepartners
reaction (cf.partners
Table 1).(cf. Table 1).
The
The data
data asasshown
shown in inTable
Table22clearly
clearlyshowshowthatthatSafranine
SafranineOOundergoes
undergoesaadiffusion
diffusioncontrolled
controlled
electron
electron transfer processprocesswhen
when∆G ∆Get is is significantly
et significantly negative,
negative, a known
a known behaviorbehavior for bimolecular
for bimolecular photo-
photo-induced
induced electron electron
transfertransfer
[46]. The[46]. The comparable
comparable behaviorbehavior of Rose
of Rose Bengal is Bengal is less conventional
less conventional since Rose
since
BengalRose Bengal
exhibits exhibits
rather lowrather low quenching
quenching rate constantsrate constants even at significantly
even at significantly negative ∆G negative ∆Gas
et values, et
values,
depictedas depicted
in Figurein7Figurefor its7triplet
for its triplet state.specific
state. This This specific behavior
behavior of Rose
of Rose Bengal
Bengal cancan
be be attributed
attributed to
to relatively
relatively lowlow electronic
electronic coupling
coupling between
between thethe Rose
Rose Bengal
Bengal excited
excited states
states andandthe the organo-borate
organo-borate salt
salt [47,48].
[47,48].

Figure7.7.Quenching
Figure Quenchingraterateconstant
constantk kTqTofofRB
RB(filled
(filledsymbols)
symbols) and
and ofof
SFSF (open
(open symbols)
symbols) as aasfunction
a function
of
q
of ∆G T for different borate salts [35].
∆Get T for different borate salts [35].
et

It has been widely reported in the literature that the efficiency of the photo-polymerization
It has been widely reported in the literature that the efficiency of the photo-polymerization process
process is directly proportional to the excited state reactivity and that a good mechanistic
is directly proportional to the excited state reactivity and that a good mechanistic understanding of this
understanding of this reactivity in solution allows the stringent interpretation of the polymerization
reactivity in solution allows the stringent interpretation of the polymerization data. These experiments,
data. These experiments, performed in low viscous solutions, however, rarely take into account the
performed in low viscous solutions, however, rarely take into account the specific features of such
specific features of such photo-polymerizing resin. The rate constants of the monitored reaction in
photo-polymerizing resin. The rate constants of the monitored reaction in solution have to be corrected
solution have to be corrected for the diffusion of the monomer in the medium by taking into account
for the diffusion of the monomer in the medium by taking into account the viscosity η of the respective
the viscosity η of the respective resin and calculating the diffusion rate constant from a simplified
resin and calculating the diffusion rate constant from a simplified Stokes-Einstein equation:
Stokes-Einstein equation:
8 ·8R⋅ R
· T⋅ T
kd = (8)
kd = 3·η (8)
3⋅ η
The value of the material described herein is kd = 4.2 × 106 M−1·s−1. The current quenching rate
constants kr under such conditions can be evaluated from Equation (9):

kd ⋅ kq
kr = (9)
k −d + k q
With k−d expressing the separation rate constant in the medium which was set equal to kd [45–
50]. As a consequence, the rate constant of the reaction between the organo-borate salts and excited
Polymers 2017, 9, 472 14 of 35

The value of the material described herein is kd = 4.2 × 106 M−1 ·s−1 . The current quenching rate
constants kr under such conditions can be evaluated from Equation (9):

kd · kq
kr = (9)
k −d + k q

With k−d expressing the separation rate constant in the medium which was set equal to kd [45–50].
As a consequence, the rate constant of the reaction between the organo-borate salts and excited dyes
will be limited by diffusion for the Safranine O/organo-borate systems. In the case of the Rose
Bengal/organo-borate systems, the quenching rates constant are limited by diffusion for the singlet
state but still below the diffusion limit for the triplet state.
Apart from the correction for the diffusion in the described medium, it has to be taken into
account that the quantum yield for the formation of the alkyl radical φP additionally depends on the
photo-physical properties of the excited states, the quenching rate constants kr and the concentration
of quencher [Q]. φR can be thus expressed as the sum of the contributions from both the singlet state
φP S and the triplet state φR T , respectively. φR S only depends on the singlet state quenching efficiency
of the dye by the organo-borate as:

1
φSR = 1 − (10)
1 + kSr · τ0S · [ Q]

where τ0 S expresses the lifetime of the singlet excited state of the dye in the absence of a quencher.
Similarly, φR T depends on the triplet state quenching efficiency and the quantum yield of the actual
triplet state φT . !
T 1
φR = φT · 1 − (11)
1 + krT · τ0T · [ Q]

With τ0 T as the lifetime of the actual triplet excited state. φT moreover is given by:

φ0T
φT = (12)
1 + kSr · τ0S · [ Q]

φT 0 being the triplet state quantum yield in the absence of a quencher. Therefore, the overall quantum
yield φR of alkyl radical formation is expressed by:
! ! !
1 φT 1
φR = 1− + 1− (13)
1 + kr · τ0S · [ Q]
S 1 + kr · τ0S · [ Q]
S 1 + kr · τ0T · [ Q]
T

The values of φR are shown in Table 3. It can be deduced from these data that the total quantum
yield of the product formation φR for the photo-initiating systems based on Rose Bengal/organo-borate
salt increases with Eox as a consequence of the rather low values of the quenching rate constants. In case
of Safranine O, the φR values are similar to each independent of the organo-borate salt used, since the
quenching rate constants kr in the resin are diffusion limited.
Figure 8 exhibit that the rate of polymerization for both Rose Bengal and Safranine O fit
a relationship with φR . These data are consistent with the results obtained from the RT-FTIR
experiments where the photo-initiating system consisting of Rose Bengal/organo-borate 4 is the most
efficient system. From these data, the efficiency of the systems based on Safranine O/organo-borate
salts is in the diffusion limit and their reactivity is not proportional to the photo-induced electron
transfer process.
Polymers 2017, 9, 472 15 of 35

Table 3. Total radical quantum yield φR formed by reaction of 1.5 × 10−2 M of borate salts and Rose
Bengal or Safranine O excited states in the photo-polymerizing resin [35]. In brackets: maximum
refractive index modulation as obtained in holography recording.

Borate RB φR SF φR
1 0.376 0.285
2 0.323 0.284
3 - 0.285
4 0.201 (0.0078) 0.285 (0.0078)
5 0.227 (0.0090) 0.284 (0.0085)
Polymers 2017, 9, 472
6 0.154 (0.0065) 0.284 15 of 35

Figure
Figure 8. R8.p Rversus
p versus φR in the photo-polymerizing resin [35].
φR in the photo-polymerizing resin [35].

4.4. Holographic Recording

4.4. Holographic Recording
The photopolymer formulations of Rose Bengal and Safranine O with respective borates were
The photopolymer
investigated formulations
in a holographic of Rose
experiment. Bengal and
Unfortunately Safranine
not O with respective
all the dye/organo-borate borates were
combinations
investigated in aphoto-initiators
form stable holographic experiment.
due to their Unfortunately
tendency to undergo not alla the dye/organo-borate
spontaneous self- or darkcombinations
reaction.
form For
stable
the photo-initiators due toholographic
suitable photo-initiators their tendencyplaneto undergo
wave gratinga recordings
spontaneous wereself- or darkusing
performed reaction.
the dye/organo-borate photo-initiating system in a photopolymer composed
For the suitable photo-initiators holographic plane wave grating recordings were performed using theof 25% SR 349, 0.1% dye,
and 1.0% organo-borate
dye/organo-borate in a matrix
photo-initiating [13,18].
system in aFigure 9 and Table
photopolymer 3 displayofthe
composed 25% refractive
SR 349, index
0.1% dye,
modulation ∆n 1 of the holograms obtained for stable photopolymers as a function of the irradiation
and 1.0% organo-borate in a matrix [13,18]. Figure 9 and Table 3 display the refractive index modulation
dose per arm of the 2-beam plane wave recording setup. Note high ∆n1 values indicate high
∆n1 of the holograms obtained for stable photopolymers as a function of the irradiation dose per arm
brightness upon vHOE reconstruction, provided over-modulation has not occurred or cannot occur
of the 2-beam plane wave recording setup. Note high ∆n1 values indicate high brightness upon vHOE
as for reflection holograms. Recording was done in reflection mode with green laser light of 532 nm.
reconstruction,
Experimental provided
details ofover-modulation
the recording setup has and
not occurred or cannot
data extraction occur as elsewhere
are described for reflection
[17].holograms.
Good
Recording wasand
∆n1 values done highin sensitivity
reflectionwere mode withfor
found green laser light
the systems based ofon532 nm. Experimental
Safranine O/organo-borate details
5 andof the
recording setup and data extraction are described elsewhere [17]. Good ∆n values
Safranine O/organo-borate 4. For the Rose Bengal based systems, also good1 ∆n1 values and medium and high sensitivity
were sensitivity
found for wasthe systems
found when based onorgano-borates
using Safranine O/organo-borate
4 and 5, good ∆n51 and Safranine
values and poorO/organo-borate
sensitivity was 4.
found
For the Rose forBengal
organo-borate 6.
based systems, also good ∆n1 values and medium sensitivity was found when
As displayed,4 aand
using organo-borates good ∆n1 dose
low5,irradiation values is and
required
pooronly to create
sensitivity a high
was found 1 in the Safranine
∆nfor organo-borate 6.
O/organo-borate 4 and Safranine O/organo-borate 5 systems as a
As displayed, a low irradiation dose is required only to create a high ∆n1 in the result of the high radical quantum
Safranine
yield φR created and the high Rp values. A high ∆n1 is also obtained for Rose Bengal/organo-borate 4
O/organo-borate 4 and Safranine O/organo-borate 5 systems as a result of the high radical quantum
and Rose Bengal/organo-borate 5, although in this case, a longer exposure time is necessary due to
yield φR created and the high Rp values. A high ∆n1 is also obtained for Rose Bengal/organo-borate 4
the lower Rp value. Finally, for the Rose Bengal/organo-borate 6 system, a moderate ∆n1 is obtained
and Rose Bengal/organo-borate
at high exposure dosage, due to5,the although
very lowinRpthis
andcase, a longer
low radical exposure
quantum yieldtime is necessarythe
φR. Surprisingly, due to
the lower R
sequence value. Finally, for the Rose Bengal/organo-borate 6 system, a moderate
p of increasing inhibition dosages (the dosage needed to start any holographic recording) ∆n 1 is obtained
for
at high exposure
each dosage,
dye exactly followsduethe to the very
sequence low Rp and
of decreasing low radical
Rp values for each quantum yield φR .couple.
dye/organo-borate Surprisingly,
By
the sequence of increasing
taking e.g., a value of ∆n inhibition
1 = 0.004, dosages (the dosagetoneeded
which corresponds to start
half of the any holographic
maximum value obtained recording)
in
holographic
for each dye exactlyrecording,
followsFigure 10 correlates
the sequence the corresponding
of decreasing Rp values dosage per arm
for each with φR in a viscous
dye/organo-borate couple.
medium.
By taking e.g.,Thus,
a valuethe of
results
∆n1 =of0.004,
the holographic recording as
which corresponds todescribed
half of the quite well correlate
maximum valuewith the in
obtained
photo-polymerization experiments.
holographic recording, Figure 10 correlates the corresponding dosage per arm with φR in a viscous
Polymers 2017, 9, 472 16 of 35

medium. Thus, the results of the holographic recording as described quite well correlate with the
photo-polymerization
Polymers 2017,
Polymers 2017, 9, 472 experiments.
9, 472 16 of
16 of 35
35

Figure 9. ∆n
Figure
Figure 9. ∆n
9. as
1∆n a function
1 as a functionof
ofthe
the irradiation dosageper
irradiation dosage per arm
arm [35].
[35].
1 as a function of the irradiation dosage per arm [35].

Figure 10. Dosage per arm necessary to obtain a ∆n1 = 0.004 as a function of φR [35].
Figure 10. Dosage
Figure 10. Dosage perper
armarm necessarytotoobtain
necessary obtain aa ∆n
∆n1 ==0.004 as a function of φR [35].
1 0.004 as a function of φR [35].

The ability
The ability to
to predict
predict and
and interpret
interpret the
the performance
performance of of aa photo
photo reactive
reactive polymer
polymer resin
resin containing
containing
a
The type II
ability
a type photo-initiating
to predict and
II photo-initiating system from
interpret
system fromthethe reactivity
theperformance of the excited
of aexcited
reactivity of the states
photo states of
reactive the involved
polymer
of the involved dyes
resin and
containing
dyes and
from
a typefrom the intrinsic
II photo-initiating properties of the
systemoffrom
the intrinsic properties photo-polymerizing medium
the reactivity of themedium
the photo-polymerizing excitedhashas been
states demonstrated.
beenofdemonstrated.
the involved dyes and from
the intrinsic properties of the photo-polymerizing medium has been demonstrated.
5. Full
5. Full Non-Local
Non-Local Reaction
Reaction Driven
Driven Diffusion
Diffusion Model
Model for
for Bayfol
Bayfol® HX
® HX Film
Film
5. Full Non-Local
We have
We Reaction
have shown
shown in Driven
in the
the Diffusion
previous
previous Section 33Model
Section that for
that the
the Bayfol®process
recording
recording HX Film
process in Bayfol
in HX film
Bayfol®® HX film can
can be
be
successfully described by a simple non-local reaction diffusion model. Having investigated in Section
successfully described byprevious
a simple non-local reaction diffusion model. Havingininvestigated ® HX in Section
We
4
have
the
shown
details of
in the
the photo-initiation
Section
mechanism
3 thatwe the recording
are now able
process
to extend
Bayfol
our simple
film can be
non-local
4 the details of the photo-initiation mechanism we are now able to extend our simple non-local
successfully
reactiondescribed
diffusion by a simple
model to aanon-local reaction
fully detailed
detailed diffusion
model. By thismodel.
this we areHavingableinvestigated
are able eliminateinmany
to eliminate Section 4
many
reaction diffusion model to fully model. By we to
the details of the photo-initiation
simplifications and and shortcomings
shortcomings of mechanism
of the
the earlier we are
earlier model. now
model. In able
In [51]
[51] we to extend
we have
have shown our
shown how simple
how the non-local
the detailed
detailed photo- reaction
photo-
simplifications
diffusion model
initiation
initiation to athe
process,
process, fully
the chain
chain detailed
growth model.
growth and By thisprocesses,
and termination
termination we are able
processes, to eliminate
the inhibition
the inhibition effectsmany
effects due to
due simplifications
to oxygen
oxygen andand
the mass
and shortcomings transport
of of
the all relevant
earlier mobile
model. Inspecies
[51] weincluding
have the
shown counter
how flow
the mass transport of all relevant mobile species including the counter flow of a non-polymerizable the of a non-polymerizable
detailed photo-initiation
component,
process, the chainto
component, togrowth
guarantee
guarantee and incompressibility can be
termination processes,
incompressibility can be incorporated.
incorporated.
the inhibition We will
We will firstly
effects due
firstly review this work
to oxygen
review this work given
andgiven
the mass
in
transport [51]
in [51] to
of to some
allsome extend
relevant to
mobile
extend rationalize later
species including
to rationalize the photopolymer formulation
the counter formulation
later the photopolymer optimization
flow of a non-polymerizable for a
optimization for a tunable tunable
component,
index modulation
modulation ∆n ∆n11 and
and mechanical
mechanical stiffness
stiffness (measured
(measured as
index
to guarantee incompressibility can be incorporated. Weas storage shear
storage
will shear modulus
firstly modulus G
review thisGphoto
photo cured).
work cured).given in [51]

to some extend to rationalize later the photopolymer formulation optimization for a tunable index
modulation ∆n1 and mechanical stiffness (measured as storage shear modulus Gphoto cured ).
Polymers 2017, 9, 472 17 of 35

5.1. Outline of the Fully Detailed Model

As outlined in [51] the description of the photo-initiation process is the starting point. First
a photon is absorbed by the dye A. A will then be converted to its excited triplet state Dye* with a rate
constant ka . The dye A can be recovered from its excited triplet state Dye* with a decay constant kr . kr
is the inverse of the lifetime τr of the excited triplet state Dye*.

ka
A −→ Dye* (14)

r k
Dye* −→ A (15)

For the formation of the primary radical molecule R• the excited triplet state Dye* reacts with
a co-initiator CI with a rate constant kd . Besides R• a protonated dye molecule HDye is formed too.

k
Dye* + CI d
−→ Dye•− + R• + H + −→ HDye + R• (16)

A finally bleached dye is generated by the reaction of the protonated dye molecule HDye with the
co-initiator CI. An intermediate co-initiator state CIint is also formed.

k
HDye + R• b
−→ bleached dye + CIint (17)

For the initiation of a chain initiator molecule M1 • the primary radicals R• react with a monomer u
with a rate constant ki . There may not be only a single type of monomer present but a mixture
of monomers with for example different functionalities could be used. Monomers with higher
functionality will form an interpenetrating network with the matrix network. Different monomer
types are not further distinguished from the chain initiator state on. Also all mobility of the resulting
chain and network molecules is assumed to be lost from this state on within our model.
k
u + R• i
−→ M1• (18)

For monomers with functionalities f > 1 it might be reasonable to assume that the initiation rate
constant ki scales with f.
After the initiation process the chain propagation has to be described by the rate constant kp .

kp
u + Mn• −→ Mn•+1 (19)

The driving force for monomer diffusion is generated by the assumed local immobilization of the
polymerized monomers. Therefore it might be reasonable to assume that the propagation rate constant
kp is independent of the functionality of the monomer type. If several types of monomers are used
they should have the same chemical type of functional group (e.g., an acrylate group) to keep the latter
assumption valid.
After chain growth chain termination has to be described. Combination or disproportionation is
a consequence of bimolecular termination as depicted below.

• k
Mm + Mn• tc
−→ Mm + n (20)

• k
Mm + Mn• td
−→ Mm + Mn (21)

It is common to combine the two respective rate constants in one bimolecular termination rate
constant kt :
k t = k tc + k td (22)
Polymers 2017, 9, 472 18 of 35

A terminated polymer is also achieved by the termination of a growing chain radical with
a primary radical.
k tp
R• + Mn• −→ RMn (23)

Primary radical recombination will be neglected in this study. For high recording powers
especially in the case of mono pulse recording (~ns pulse time) this assumption has to be revised again.
Besides initiation, growth and termination the inhibition as a non-ideal effect has to be included.
For free radical polymerization oxygen (in the following denoted as Z) usually governs inhibition.
Oxygene (Z) is able to react with the excited triplet state of the dye, the primary radical or the
macro radical.
k z1
Dye* + Z −→ oxydized leuco dye (24)
k
R• + Z z2
−→ scavengend radical (25)
k
M• + Z
z3
−→ terminated polymer (26)

The following assumptions can be made considering the mass transport of the different mobile
species in the photopolymer. Z may diffuse from the plastic substrate into the photopolymer at a rate
proportional to–(Z–Z0 ). Z0 is denoted as the equilibrium concentration of oxygen in the photopolymer
layer. Within the photopolymer layer Z diffuses too. Mobile components are considered to be
monomers u, the primary radical R• and all other components of the photo-initiator system. Within
this model we consider as immobile components all chain radicals (they may be also cross linked in
case of monomer functionalities f > 1) and the cross linked matrix. A non-reactive contrast agent B is
introduced to compensate the mass flow of monomers from the dark fringe towards the bright fringe
during the holographic recording and by this equalizing the osmotic pressure. By the counter-diffusion
of the contrast agent B to the monomers u the incompressibility is given at any time at any location.
The contributions to incompressibility violation by the initiator system components are neglected as
they are only present in very dilute concentrations. Still neglected for incompressibility violation is
volume shrinkage that occur for example in free radical polymerization of acrylic monomers. In Bayfol®
HX film photopolymers for example the volume shrinkage is below 3%, depending on the composition.
In the presence of a spatially sinusoidal variation of the intensity during recording of a plane-wave
grating the reactions described above have to be translated into a set of integro-differential equations.
In order to solve the set of integro-differential equations it is straightforward to expand them into
a Fourier series of the grating period Λ of the respective grating vector length K = 2π/Λ.
In [51] we further outlined that the rate constant ka that describes the excitation rate of the dye A
to its excited triplet state Dye* is expressed as:
0
k a = Φ · ε · d · I0 (27)
 
ln Ts f /T0
ε= (28)
A0 ( t = 0) · d
0 Ts f · I0 · λ 1 − exp[−ε · A0 (t) · d]
· Θ(t) · Θ texp − t


I0 = · (29)
NA · h · c d
where d is the traversed thickness of the light according to the recording conditions given in cm. ε
is the molar extinction coefficient given in cm2 /mol. Φ being identical to φT 0 from Section 4 is the
quantum efficiency to generate an excited triplet state Dye* per photon that traverses the photopolymer.
T0 is the transmission at the recording wavelength before illumination and Tsf is the final transmission
after recording and is set to 0.9. A0 is the initial zero-order spatial Fourier component of the dye
concentration, given in mol/cm3 . NA is the Avogadro’s constant, h the Planck’s constant and c the
vacuum speed of light. Θ denotes the Heaviside function. The average number of absorbed mol
photons per second and per unit volume through the photopolymer layer during the full recording
Polymers 2017, 9, 472 19 of 35

interval (0, texp ) is described by Equation (29). A0 (t) denotes the transient zero order Fourier component
of the dye distribution. By A0 (t) also the transient bleaching during the recording interval (0, texp ) is
taken into account within this model.
Rate equations for the extensive variables were formulated and the details are given in [51]. One
example is the rate equation for the dye A. All concentrations are given in mol/cm3 .
 
∂A ∂ ∂A
= DA − k a · (1 + V · cos(K · x )) · A + kr · Dye* (30)
∂t ∂x ∂x

Another example is the rate equation for the monomer u.

" #
( x − x 0 )2
  Z ∞
∂u ∂ ∂u • 1
= Du − ki · R · u − √ · exp − · k p · M• · u · dx 0 (31)
∂t ∂x ∂x −∞ 2π · σ2 2 · σ2

The non-local parameter is denoted as σ2 again [22]. As reported in [17], σ2 was set to 100 nm2 .
For the polymerized monomer units N the following rate equation can be formulated:
" #
( x − x 0 )2
Z ∞
∂N 1
= √ · exp − · k p · M• · u · dx 0 (32)
∂t −∞ 2π · σ2 2 · σ2

The rate equation for the non-polymerizable contrast agent B has to account for the
incompressibility and as a consequence has to be written as:
 
∂B ∂ ∂u
= −VF · Du (33)
∂t ∂x ∂x

The ratio of molar volumes of the monomer u and the contrast agent B is denoted as VF.
The zero-order Fourier component of the polymerized monomer units N0 , which is a measure of
the spatially averaged monomer conversion and by this a measure for the increasing internal viscosity,
was used in [51] to describe the slowdown of diffusion speed of all mobile components. Therefore the
diffusion coefficients D are scaled down linearly in N0 (t) via:
 
1 N0 (t)
D = D0 − 1 − · D0 · (34)
α u0 ( t = 0)

From rheological properties like the ratio of the shear modulus of the unrecorded state G0 versus
the shear modulus in the fully cured state Gphoto cured the scaling constant α can be deduced. Due to the
possible large concentrations and concentration gradients of the monomers Du has the character of
an inter-diffusion coefficient. Due to small concentrations of the initiator system components in the
photopolymer formulation DA , DHDye , DCI and DR have the character of tracer diffusion coefficients.
Further we relate the diffusion coefficients D0 of the various mobile components to that of the
monomer unit in a rubbery network (Equation (29) in [51]) by the following scaling:
 ρ 4/3
D0 ∝ (35)
M

For further simplification we set DA = DHDye and DCI = DR .

By Fourier-series expansion with respect to the basic grating period Λ the above mentioned set of
integro-differential equations under appropriate initial conditions can be solved.
With all the respective volume fraction profiles φk of the monomers u, the polymerized
monomers N, the contrast agent B and the matrix b as background and their different refractive
indices, the refractive index profile ∆n can be calculated using the Lorentz-Lorenz relation and the
Polymers 2017, 9, 472 20 of 35

incompressibility condition. In Figure 11 an example of such a ∆n profile is depicted with the inhibition
period clearly visible.
Polymers 2017, 9, 472 20 of 35

FigureFigure 11. Simulated

11. Simulated refractive
refractive index profile ∆n
indexprofile ∆n versus
versustime
timet and
t and space
space coordinate
coordinatex forx Bayfol ® HX®
for Bayfol HX
film, film, calculated
calculated for for
λ =λ633
= 633 nm,
nm, SFSF==4525 1/mm, I0I0= =
4525 1/mm, 9.12 mW/cm
9.12 mW/cm 2 and2 tand
exp = 3 s [51].
texp = 3 s [51].

In the two wave approximation outlined in Kogelnik’s work [1] for the first diffraction order of
In the
the two wave
hologram only approximation outlined
the first order Fourier in Kogelnik’s
components ∆n1 of work [1] for the
the refractive indexfirst diffraction
profile ∆n is oforder
of theinterest.
hologram only the first order Fourier components ∆n 1 of the refractive index profile
At each time t we can therefore write for ∆n1, which is then identical to the index modulation ∆n is of
interest. At each time t we
of the holographic grating: can therefore write for ∆n 1 , which is then identical to the index modulation
of the holographic grating:
Δn1 = dark2
(
n2 + 2 )
2

2 ⋅
(6n⋅ ndark
+2 )
∆n1 = 6dark ·
 ·ndark (36) (36)
 uu  n 2n2u−−11 nn22b −−11  nnb22b −−11  + φpp · n np 2p−−11 − nnb22b −−11  
2
BB  n Bn B−−11
  2  
2
φ 
· u − b
 + φ 
· − 
φ1 1⋅  2n2u +2 − n22 +2  + φ1 1⋅  2n2 +2 − n22 +2  + φ1 1⋅  2n2p +2 − n22 +2  
  nu + 2 nbb + 2   nB B+ 2 nbb + 2   n p + 2 nb + 2  
b

HereHerendarkndark
denotes
denotes the refractiveindex
the refractive index of unrecorded
of the the unrecorded photopolymer,
photopolymer, nu the index
nu the refractive refractive
indexofofthe
themonomer,
monomer, nB nthe
B the refractive
refractive index index
of theof the contrast
contrast agent,
agent, nN the nN refractive
the refractive indexindex
of theof the
polymerized
polymerizedmonomersmonomers andandnbnbthe therefractive
refractive indexindex ofof the thematrix.
matrix.These These refractive
refractive indices
indices were were
measured independently
measured independently in our case. φ1 in our case. φ k
1 k is the first-order Fourier component of the volume-fraction
the first-order Fourier component of the volume-fraction
profile φk ofφthe
profile k of the component k of the photopolymer. Extension to a multi-component monomer
component k of the photopolymer. Extension to a multi-component monomer mixture
mixture is straightforward.
is straightforward.
5.2. Estimation
5.2. Estimation of the
of the Model
Model Parametersby
Parameters byComparing
Comparing Experimental
Experimentaland Simulated
and Dosage
Simulated Response
Dosage Response Curves
Curves of RGB Sensitive Bayfol® HX Film
of RGB Sensitive Bayfol® HX Film
In [51] we described how we estimated the parameters for the model described above by fitting
In [51] we described how we estimated the parameters for the model described above by fitting
the model to the measured R, G and B dosage-response curves of a specific Bayfol® HX film grade at
the model to the measured
two different R, G and B Independently
spatial frequencies. dosage-response curves of
measured a specific
refractive Bayfolof® HX
indices all film grade at
relevant
two different spatial frequencies. Independently measured refractive indices
components were used. Also experimental data for kp [52] and for kd and kr for the formulationof all relevant components
werecomponents
used. Also were
experimental
set as starting for kpfor
datavalues [52]
theand for kd andkkd rand
fit procedure. forkthe formulation
r were deduced from components
respective were
set asexperiments
starting values for theinfit
as described procedure.
Section kd and the
4. For example kr were deduced
lifetime τr of the from
excitedrespective
triplet stateexperiments
of the dye as
was found
described to be of
in Section 4. the
Fororder
exampleof 10the
μs,lifetime
thereforeτrkrof
is the
of the ordertriplet
excited of 1/10state
μs−1. of
According
the dye to wasSection
found4 to be
of theone can of
order identify
10 µs,ktherefore kr is of the order of 1/10 µs−1accessible
q = kd·CI0 (t = 0) from the experimentally
. According quenching
to Sectionconstant
4 onekqcan of the
identify
excited triplet state Dye *. As the whole set of rate equations can be scaled by an arbitrary factor in
kq = kd ·CI0 (t = 0) from the experimentally accessible quenching constant kq of the excited triplet state
time it is important to know at least a few of the above mentioned rate constants. This then leads to
Dye* . As the whole set of rate equations can be scaled by an arbitrary factor in time it is important
a consistent set of physically reasonable model parameters. In Figure 12 the obtained results are
to know at least a few of the above mentioned rate constants. This then leads to a consistent set of
shown.
physicallyAfter
reasonable
parameter model
value parameters.
optimization In Figure 12 the obtained
the simulations show a results are shown.
very good fit quality to the
After parameter value optimization the simulations show
experiments. The good fit quality is underlined by the fact that the parameters for a very good fit quality to the
the polymerization
experiments. The good fit quality is underlined by the fact that the parameters for the polymerization
Polymers 2017, 9, 472 21 of 35
Polymers 2017, 9, 472 21 of 35

(ki , kp(k, ik, tk,p,kktpt,,kktpz2 kz3,, σσ22) and

, k,z2k, z3 andmass
masstransport
transportof monomer
of monomer and oxygen (Du, α,(D
and oxygen DZu,, Zα, D[51])
0, τ Z , Z0were kept were
, τ [51])
kept almost
almostidentical identicalininallall6 6cases cases(3(3different λ and
different λ and 2 different
2 different SF)SF)
depicted in Figure
depicted in Figure12. In
12.addition
In addition
parameters related to the initiator system (φ,k k,r,kkd,, kkb,, D A, DCI) could be kept fixed for the different SF
parameters related to the initiator system (φ, r d b DA , DCI ) could be kept fixed for the different SF
for reflection and transmission geometry for each recording wavelength. Only for the blue recording
for reflection and transmission geometry for each recording wavelength. Only for the blue recording
in transmission geometry φ, kr, kd, kb and ki had to be chosen significantly different from the other
in transmission geometry φ, kr , kd , kb and ki had to be chosen significantly different from the other
cases (~50%).
cases (~50%).

(a) (b)
Figure 12. Experimental and simulated dosage-response curves for RGB sensitive Bayfol® HX film;
Figure 12. Experimental and simulated dosage-response curves for RGB sensitive Bayfol® HX film;
(a) reflection geometry data; (b) transmission geometry data [51].
(a) reflection geometry data; (b) transmission geometry data [51].
In sum we can state that the fully detailed non-local reaction-diffusion model is able to describe
In
oursum we can state
photopolymer that the
recording fully detailed
behavior very wellnon-local
in a wide reaction-diffusion model is able
variety of recording conditions to describe
and also in
wide variations of spatial frequencies.
our photopolymer recording behavior very well in a wide variety of recording conditions and also in
wide variations of spatial frequencies.
5.3. Improving the Spatial Frequency Response Based on the Fully Detailed Reaction-Diffusion Model
5.3. Improving
As wethe Spatialpreviously
reported FrequencyinResponse
Section 3,Based on the Fully of
as a consequence Detailed
the lowReaction-Diffusion Model∆n1
spatial frequency cutoff,
for transmission holograms in saturation recording is only about half the value of ∆n1 for reflection
As we reported previously in Section 3, as a consequence of the low spatial frequency cutoff, ∆n1
holograms at comparable power densities. This fact can be easily rationalized by the picture that at
for transmission holograms in saturation recording is only about half the value of ∆n1 for reflection
large grating spacing Λ (low spatial frequencies SF) the diffusing monomers already react on their
holograms at comparable power densities. This fact can be easily rationalized by the picture that at
travel to the bright fringe before reaching the bright fringe’s center. By that higher harmonic orders
largeingrating spacing
the refractive Λ (low
index spatial
profile are built up. SF)
∆n(x, t)frequencies the diffusing
Therefore only partmonomers already
of the complete react
dynamic on their
range
traveloftothe
thephotopolymer
bright fringeisbefore reaching the bright fringe’s center. By that higher
transformed into the desired first Fourier component ∆n1. From the modelharmonic orders in
the refractive index profile ∆n(x, t) are built up. Therefore only part of the
predictions we can expect that increasing the monomer mobility, by changing the composition of complete dynamic range
of thephotopolymer
photopolymer formulation should improve
is transformed into thethe ∆n1 at large
desired component ∆n1 . From the model
grating spacing.
first Fourier
predictions Thewemonomers
can expect can that
be selected as a the
increasing mixture of monofunctional
monomer mobility, byand multifunctional
changing species.
the composition of
Upon photo-polymerization the multifunctional species
photopolymer formulation should improve the ∆n1 at large grating spacing. together with the monofunctional species
generate a rigid cross linked polymer network. Multifunctional monomers will generate a very rigid
The monomers can be selected as a mixture of monofunctional and multifunctional species. Upon
network that gives the hologram long term stability. A too high amount of multifunctional species
photo-polymerization the multifunctional species together with the monofunctional species generate
however vitrifies this photopolymer network already at low monomer conversion and decreases
a rigid cross linked
monomer polymer
mobility network.
at an early stage.Multifunctional
On the other hand monomers will generate
a high amount of small,amonofunctional
very rigid network
that gives the hologram long term stability. A too high amount of multifunctional
monomers generates flexible linear segments in the network and allows the monomer mobility speciesto however
be
vitrifies this photopolymer network already
kept high even at elevated monomer conversion. at low monomer conversion and decreases monomer
mobility Different
at an early stage. On the
photopolymer other hand
formulations werea evaluated
high amount in redoflaser
small,
(633monofunctional monomers
nm) reflection hologram
recording
generates to test
flexible this assumption
linear segments in [51].
theThe total weight
network fractionthe
and allows of amonomer
monomer mobility
mixture that consists
to be kept high
of a mono- and a multifunctional
even at elevated monomer conversion. monomer species was kept constant in these formulations. To assure
that any effect
Different on ∆n1 caused
photopolymer by varying thewere
formulations monomer mixing
evaluated inratio
red can
laserbe(633
attributed to the change
nm) reflection in
hologram
monomer mobility, the monofunctional and the multifunctional monomers were chosen such that
recording to test this assumption [51]. The total weight fraction of a monomer mixture that consists of
they have almost the same double-bond density per volume (meaning also that the molecular weight
a mono- and a multifunctional monomer species was kept constant in these formulations. To assure
that any effect on ∆n1 caused by varying the monomer mixing ratio can be attributed to the change in
monomer mobility, the monofunctional and the multifunctional monomers were chosen such that they
have almost the same double-bond density per volume (meaning also that the molecular weight of
Polymers 2017, 9, 472 22 of 35
Polymers 2017, 9, 472 22 of 35

the monofunctional
of the monofunctional monomer is proportionately
monomer smallersmaller
is proportionately than that
thanof that
the multifunctional monomer),
of the multifunctional
the same chemical type of double bond (acrylate) and almost identical refractive indices.
monomer), the same chemical type of double bond (acrylate) and almost identical refractive indices.
As a As a result
result of this
of this experimental
experimental testtest
thethe
index modulation∆n
indexmodulation ∆n11 versus
versusthethemonomer
monomer mixing
mixingratio
ratio is
is depicted
depicted (right y-axis)
(right y-axis) in Figure
in Figure 13. In addition
13. In addition to represent
to represent the mechanical
the mechanical rigidity
rigidity the shear
the shear modulus
moduluswith
G, measured G, measured
an optical with an optical(Physica
Rheometer Rheometer (Physica
MCR MCR 301,
301, method methodindescribe
describe in [53]) for
[53]) is shown is the
shown for the unrecorded state as G0 and for the homogeneously photo cured state as Gphoto cured (left
unrecorded state as G0 and for the homogeneously photo cured state as Gphoto cured (left y-axis). It is
y-axis). It is very obvious that if the multifunctional monomers are the majority component ∆n1 drops
very obvious that if the multifunctional monomers are the majority component ∆n1 drops significantly.
significantly. This drop in ∆n1 by increasing amount of multifunctional monomers can be attributed
This drop in ∆n by increasing amount of multifunctional monomers can be attributed to the early
to the early1 densification of the photopolymer network and therefore strong attenuation of the
densification
monomerofdiffusion
the photopolymer network
constant. Note that Gand therefore strong attenuation of the monomer diffusion
0 does not change with the monomer mixing ratio. This
constant. Note that G does
suggests identical rheological
0 not change with
conditions for the themonomers
monomertomixing
diffuseratio.
in the This
very suggests identical
initial state of
rheological conditions for the monomers to diffuse in the very initial state of exposure.
exposure.

Figure 13. ∆n
13. ∆n 1 (right y-axis, blue circles) and shear modulus G (left y-axis), before (open squares) and
Figure 1 (right y-axis, blue circles) and shear modulus G (left y-axis), before (open squares) and
after flood cure (grey squares), versus monomer mixing ratio (fraction of the multifunctional
after flood cure (grey squares), versus monomer mixing ratio (fraction of the multifunctional monomer
monomer increasing from 0 to 1) [51].
increasing from 0 to 1) [51].
Reflection hologram recording shows a high spatial frequency SF. Therefore the diffusion time
Reflection
scale τD in hologram recording
the early stage shows is
of recording a high spatial
still small. frequency
This holds even SF.for
Therefore the diffusion
the samples with
time multifunctional
scale τD in themonomer species
early stage of only. The results
recording in Figure
is still small.13 This
therefore
holdssuggest
eventhat
forinthe
thesamples
reflection with
recording case
multifunctional most of the
monomer droponly.
species in ∆n1The
can results
be attributed to different
in Figure diffusion
13 therefore attenuation
suggest that indue
thetoreflection
the
different emerging photopolymer network densities (rigidities) at different monomer
recording case most of the drop in ∆n1 can be attributed to different diffusion attenuation due to themixing ratios.
From the experimental results described above we decided for subsequent simulations to
different emerging photopolymer network densities (rigidities) at different monomer mixing ratios.
identify the diffusion attenuation factor α as:
From the experimental results described above we decided for subsequent simulations to identify
the diffusion attenuation factor α as: G photo cure d
α = G photocured (37)
α= G0 (37)
G0
As we now know the relation of the diffusion attenuation factor α versus the mixing ratio of
As we now know the relation of the diffusion attenuation factor α versus the mixing ratio of
mono- and multifunctional monomers via the relation shown in Figure 13 we simulated the spatial
mono- and multifunctional monomers via the relation shown in Figure 13 we simulated the spatial
frequency response of a standard formulation versus a formulation with a majority of
frequency response of a standard formulation versus a formulation with a majority of monofunctional
monofunctional monomers for red laser (633 nm) saturation recording. We used all the other model
monomers for red
parameters laser (633by
we identified nm) saturation
fitting recording.
the dosage We used
response curves for all thethis.
doing other model parameters we
identifiedThe
by fitting
model the dosage
almost responsepredicts
quantitatively curves for
the doing this. values as can be clearly seen from
experimental
The model
Figure almost
14. This quantitatively
proves the suitabilitypredicts the experimental
of the model for formulationvalues as can By
optimization. be implementing
clearly seen from
this fully detailed reaction-diffusion model for Bayfol ® HX film, we identified and rationalized a
Figure 14. This proves the suitability of the model for formulation optimization. By implementing
formulation
this fully detaileddesign strategy to tune model
reaction-diffusion spatial frequency ® HX film,
for Bayfolresponse of thewephotopolymer to a rationalized
identified and desired
property profile.
a formulation design strategy to tune spatial frequency response of the photopolymer to a desired
property profile.
Polymers 2017, 9, 472 23 of 35
Polymers 2017, 9, 472 23 of 35

Polymers 2017, 9, 472 23 of 35

Figure
Figure 14. 14. Experimental
Experimental values
values of ∆n(symbols)
of ∆n 1 (symbols) and simulated ∆n1 response versus spatial frequency
and simulated ∆n1 response versus spatial frequency
1
SF for a standard formulation (solid line) and a formulation with a majority of monofunctional
SF for a standard formulation (solid line) and a formulation with a majority of monofunctional
components (dashed line) [51]. The typical error of experimentally determined ∆n1 values is about
components
±0.001.
(dashed line) [51]. The typical error of experimentally determined ∆n1 values is
about ±0.001.
Figure 14. Experimental
6. Mechanical Modulus from values
Softof Rubber
∆n1 (symbols) and Rubber
to Hard simulated ∆n1 response versus spatial frequency
SF for a standard formulation (solid line) and a formulation with a majority of monofunctional
6. Mechanical Modulus
As can fromfrom
be imagined Soft
theRubber
previousto Hardthe
section Rubber
modulus G0 in the unrecorded state should be
components (dashed line) [51]. The typical error of experimentally determined ∆n1 values is about
reasonably large to guarantee sufficient mechanical stability for recording and web-coating [53] and
As can ±0.001.
be imagined from the previous section the modulus G0 in the unrecorded state should
also α should stay close to 1 to allow a fast diffusion of monomers until high conversion has been
be reasonably
achieved large
6. Mechanicaland by tothis
guarantee
Modulus a high
from
sufficient
∆n1Soft
overRubber mechanical
wide rangeto Hardof SF canstability
Rubber be realized for(see
recording
Figure 14).andAlsoweb-coating
a small G0 [53]
and alsoinαthe should stay close to 1 to allow a fast diffusion of monomers until
unrecorded state gives an inherent tack that allows easy lamination to the recording stack with high conversion has been
As can be imagined from the previous section the® modulus G0 in the unrecorded state should be
achievedexcellent
and byoptical
this a and ∆n1 over coupling.
highmechanical wide range of SF can
In Bayfol be realized
HX type photopolymers(see Figure
G0 is 14). Also
usually in a small G0
the
reasonably large to guarantee sufficient mechanical stability for recording and web-coating [53] and
range of 0.1 to
in the unrecorded 0.5 MPa. This corresponds to the property profile of a soft rubber.
also α should stay close to 1 to allow a fast diffusion of monomers until high conversion has been with
state gives an inherent tack that allows easy lamination to the recording stack
However in certain applications the mechanical modulus ® HX typeGphoto cured in the fully recorded or photo-
excellentachieved
optical and
and bymechanical
this a high ∆ncoupling. In Bayfol
1 over wide range of SF can be realizedphotopolymers G0 isa small
(see Figure 14). Also usually G0 in the
cured should be as high as possible for example to better sustain high pressures and/or high
range oftemperatures
in0.1
theto 0.5 MPa.
unrecorded This
state corresponds
gives an inherenttotackthe property
that
during vHOE integration into the final device.
profile
allows easy of
lamination a soft rubber.
to the recording stack with
As we have shown already in Figure 14,
excellentin
However optical
certain andapplications
mechanical coupling. In Bayfol® HX type photopolymers
Gphoto G0 the
is usually in the
by adapting the mixing ratio of mono-the andmechanical
multifunctional modulus
monomers α cancured in
be varied by fully recorded
2.5 decades. or
range of 0.1 to 0.5 MPa. This corresponds to the property profile of a soft rubber.
photo-cured
By thisshould
a Gphoto be
cured as highfor
typical as very
possible for example
hard rubbers to better
or typical for soft sustain high pressures
thermoplastic and/or
materials can be high
However in certain applications the mechanical modulus Gphoto cured in the fully recorded or photo-
achieved.
temperatures However
during vHOE for integration
the Bayfol® HX intotype
the photopolymer
final device. depicted
As we in Figure
have shown14 the increaseinofFigure
already α 14,
cured should be as high as possible for example to better sustain high pressures and/or high
goes on
by adapting the the cost ofratio
mixing the ∆n 1. By using an adapted matrix [54] this loss of ∆n1 by increasing α can be
of mono- and multifunctional monomers α can be varied by 2.5 decades.
temperatures during vHOE integration into the final device. As we have shown already in Figure 14,
completely avoided as it is shown in Figure 15. In this case, we increase α by 3 decades boosting Gphoto cured
By this abyGadapting typical
the
photo cured mixing for ratiovery hardand
of mono- rubbers or typical
multifunctional for softα can
monomers thermoplastic
be varied by 2.5materials
decades. can be
to >100 MPa without any significant ® HX
variation of ∆n1.
achieved. By However
this a Gphoto cured
for typical for very
the Bayfol hard rubbers
type or typical fordepicted
photopolymer soft thermoplastic
in Figurematerials can be
14 the increase of α
achieved. However for the Bayfol ® HX type photopolymer depicted in Figure 14 the increase of α
goes on the cost of the ∆n1 . By using an adapted matrix [54] this loss of ∆n1 by increasing α can be
goes on the cost of the ∆n1. By using an adapted matrix [54] this loss of ∆n1 by increasing α can be
completely avoided as it is shown in Figure 15. In this case, we increase α by 3 decades boosting
completely avoided as it is shown in Figure 15. In this case, we increase α by 3 decades boosting Gphoto cured
Gphoto cured to >100
to >100 MPaMPa without without any significant
any significant variation variation
of ∆n1. of ∆n1 .

Figure 15. ∆n1 (right y-axis, blue circles) and shear modulus Gphoto cured after flood cure (left y-axis, grey
squares) versus monomer mixing ratio (fraction of the multifunctional monomer increasing from 0 to 1).

Figure 15. ∆n1 (right y-axis, blue circles) and shear modulus Gphoto cured after flood cure (left y-axis, grey
Figure 15. ∆n1 (right y-axis, blue circles) and shear modulus Gphoto cured after flood cure (left y-axis,
squares) versus monomer mixing ratio (fraction of the multifunctional monomer increasing from 0 to 1).
grey squares) versus monomer mixing ratio (fraction of the multifunctional monomer increasing
from 0 to 1).
Polymers 2017, 9, 472 24 of 35
Polymers 2017, 9, 472 24 of 35

7. Bleaching Process
7. Bleaching Process
After recording the vHOE it is mandatory to cure any residual writing monomers and bleach
any residual dye from
After recording the the
vHOEinitiator system. For
it is mandatory this any
to cure usually a flash
residual of UV/VIS
writing monomers lightandcanbleach
be used.any
However carefrom
residual dye has theto be taken not
initiator to useFor
system. a too
thishigh dosage
usually of UV
a flash light as there
of UV/VIS is always
light can be used.theHowever
threat to
induce
care hasyellowing
to be takeninnot theto substrate
use a too highand/or photopolymer
dosage of UV lightlayer.
as thereIn isthe laminated
always stateto oxygen
the threat induce
inhibition
yellowing of bleachingand/or
in the substrate of the residual dye is layer.
photopolymer very well
In thesuppressed
laminated and statethe necessary
oxygen dosage
inhibition for
of the
excellent
bleaching results is lowest.
of the residual dyeIn this well
is very section, we describe
suppressed and the how to optimize
necessary dosagethe forbleaching without
excellent results is
additional yellowing.
lowest. In this section, we describe how to optimize the bleaching without additional yellowing.
For this we use a combination of narrow banded UV LEDs and VIS LEDs to bleach and cure the
photopolymer. Comparedto atometal
photopolymer. Compared a metal
halidehalide bulb
bulb this this combination
combination allows for allows for homogeneous
(i) a rather (i) a rather
homogeneous
illumination; (ii)illumination,
a control of(ii)the a control of the UVand
UV wavelength wavelength and (iii)
(iii) a control of UVa control
dosage of and
UV dosage
VIS dosageand
VIS dosageFurthermore,
separately. separately. Furthermore,
the UV LED and the UV
VIS LED
LED and VIS LEDcan
combination combination
be switched canonbeand
switched on and
off as required
off as required
by the by the step-and-repeat
step-and-repeat process whereas process
such awhereas such a cycling
cycling would woulddecrease
dramatically dramatically decrease
the lifetime of
the lifetime
a metal of abulb
halide metal or halide
mercury bulb or mercury bulb.
bulb.
The spectrum
spectrum of of aa standard
standardmetal metalhalide
halidebulbbulb(DYMAX
(DYMAX 2000;
2000; VIS:VIS:
50 mW/cm
50 mW/cm 2 and 2 UV:
and
52 mW/cm 2
2) is shown in Figure 16 together with the separation into
UV: 52 mW/cm ) is shown in Figure 16 together with the separation into UV and VIS radiation UV and VIS radiation parts as
defined by the by
parts as defined sensitivity of theofdetectors
the sensitivity we we
the detectors useuse
forforintensity
intensitymeasurements.
measurements.ObviouslyObviously this
spectrum differs
differsfrom froma UV a UV LED LEDandand VIS LED
VIS LED spectrum
spectrum (see also(see also 16
Figure Figure
for the16VISforLED
the spectrum).
VIS LED
spectrum). Most prominently
Most prominently is the broader is the
UVbroader
part in UV part in theofspectrum
the spectrum the metalofhalide
the metal
bulb halide bulb well
that extent that
extent
below well below
the 365 ± 5 thenm 365line±of5 the
nm UV lineLED.
of the UV LED.

(a) (b)
Figure 16. (a)
Figure 16. (a) UV
UV light-curing
light-curing flood
flood lamp
lamp spectral
spectral distribution
distribution from
from aa standard
standard metal-halide
metal-halide bulb
bulb
(DYMAX 2000; source: www.dymax.com). Indicated are also the UV and VIS radiation
(DYMAX 2000; source: www.dymax.com). Indicated are also the UV and VIS radiation components components
as
as defined
defined bybythe
thesensitivity of of
sensitivity thethe
detectors (Gigahertz-Optik–X11
detectors (Gigahertz-Optik–X11 Optometer; UV: UV-3719-4;
Optometer; VIS:
UV: UV-3719-4;
RW-3705-4); (b) White LED Philips ClearFlood BVP650 [55].
VIS: RW-3705-4); (b) White LED Philips ClearFlood BVP650 [55].

The intensity itself and the dosage, i.e., the exposure time times the intensity, govern the
The intensity itself and the dosage, i.e., the exposure time times the intensity, govern the bleaching
bleaching result. A better bleaching result is defined as a higher transmission in the visible spectral
result. A better bleaching result is defined as a higher transmission in the visible spectral region
region (~400–800 nm) as measured in a spectrometer with the film laminated onto a glass slide. Even
(~400–800 nm) as measured in a spectrometer with the film laminated onto a glass slide. Even though
though both parameters may be carefully optimized for each® Bayfol® HX film grade, as for every
both parameters may be carefully optimized for each Bayfol HX film grade, as for every organic
organic material we find: the more intensity and dosage is applied during bleaching the better the
material we find: the more intensity and dosage is applied during bleaching the better the result.
result. However, if the UV wavelength is too short (<365 nm) and the dosage is too high yellowing of
However, if the UV wavelength is too short (<365 nm) and the dosage is too high yellowing of the
the material starts and yellowing increases with increasing UV dosage.
material starts and yellowing increases with increasing UV dosage.
This is exemplarily shown in Figure 17a in which the UV LED and VIS LED combination is
This is exemplarily shown in Figure 17a in which the UV LED and VIS LED combination is
compared to the metal halide bulb: In both cases the higher dosage yields a better transmission in the
compared to the metal halide bulb: In both cases the higher dosage yields a better transmission in the
visible region. Furthermore, it can be seen that the performance of both devices is similar if the same
visible region. Furthermore, it can be seen that the performance of both devices is similar if the same
dosage is applied.
dosage is applied.
To illustrate the yellowing effect we bleached a certain Bayfol® HX film grade with a TAC
substrate using both curing systems. As can be seen in Figure 17b the transparency in the deep blue
Polymers 2017, 9, 472 25 of 35

To illustrate the yellowing effect we bleached a certain Bayfol® HX film grade with a TAC
Polymers 2017, 9, 472 25 of 35
substrate using both curing systems. As can be seen in Figure 17b the transparency in the deep
blue (~400 nm) is decreased with increasing dosage of the metal halide bulb radiation. In contrast,
(~400 nm) is decreased with increasing dosage of the metal halide bulb radiation. In contrast, the
the transparency around 400 nm stays relatively constant if the UV LED and VIS LED is used for
transparency around 400 nm stays relatively constant if the UV LED and VIS LED is used for
bleaching. This prominent effect appears even though the applied UV dosage by the UV LED and VIS
bleaching. This prominent effect appears even though the applied UV dosage by the UV LED and
LEDLED
VIS combination is roughly
combination a factor
is roughly of 2 larger
a factor compared
of 2 larger to thetometal
compared halide
the metal bulb.bulb.
halide Moreover, it canitbe
Moreover,
clearly seen that in the case of the UV LED and VIS LED combination the higher UV
can be clearly seen that in the case of the UV LED and VIS LED combination the higher UV dosage dosage helps to
increase
helps the overall
to increase thetransmission in the visible
overall transmission in thespectral region (~400–800
visible spectral nm). nm).
region (~400–800

(a) (b)
Figure
Figure 17.
17. Comparison
Comparisonof ofbleaching
bleachingresults
resultsininwhich
whicha ametal
metalhalide
halidebulb
bulband
anda UV
a UV LED
LED and
and VIS LED
VIS LED
combination are used. (a) For the same dosage the performance of both curing systems
combination are used. (a) For the same dosage the performance of both curing systems is similar. is similar.
Moreover,
Moreover, aahigher
higherdosage
dosage usually
usuallygives a better
gives bleaching
a better result;
bleaching (b) Yellowing
result; effect:effect:
(b) Yellowing due toduethe UV-
to the
B part of the metal halide bulb (cf. Figure 16) the transparency in the deep blue (black
UV-B part of the metal halide bulb (cf. Figure 16) the transparency in the deep blue (black arrow) arrow) is is
significantly decreased. This effect is not apparent if the UV LED and VIS LED combination
significantly decreased. This effect is not apparent if the UV LED and VIS LED combination is used is used
since
since this
this bleaching
bleachingunit
unitonly
onlyhas
hasaasingle
singleUVUVline
lineradiation
radiationspectrum
spectrumatat365365± ±
5 nm
5 nm[56].
[56].

8. How to Mass Produce Volume Holographic Optical Elements with Bayfol® ®HX Film
8. How to Mass Produce Volume Holographic Optical Elements with Bayfol HX Film
For the commercialization of vHOEs simple and robust mass production schemes are of utmost
For the commercialization of vHOEs simple and robust mass production schemes are of utmost
importance. Therefore at the end of this review we would like to touch on this aspect too. In [55,56]
importance. Therefore at the end of this review we would like to touch on this aspect too. In [55,56]
we reported about a contact copy process from a so-called master vHOE. In this copy process Bayfol®
we reported about a contact copy process from a so-called master vHOE. In this copy process Bayfol®
HX film is laminated to this master vHOE. A single scanning laser line copies the phase hologram of
HX film is laminated to this master vHOE. A single scanning laser line copies the phase hologram of
the master vHOE into the copy film. The copy film is moved in a step and repeat process through the
the master vHOE into the copy film. The copy film is moved in a step and repeat process through the
replication line. To record a master vHOE in Bayfol® HX film without being bound to size limitations
replication
by line.limits
laser power To record a master in
we described vHOE Bayfol® HXprinter.
[57] ainholographic film without being bound to size limitations
by laser power limits we described in [57] a holographic printer.
First we review the main topics of computer generated volume holographic optical elements
First we
generated in areview
printerthe
andmain topics
which can beof used
computer generated
as master vHOEs. volume
Secondholographic optical
we outline the mainelements
topics
generated in a printer and which can be used as master vHOEs.
of the replication procedure indicated above for Bayfol® HX film. Second we outline the main topics of
the replication procedure indicated above for Bayfol® HX film.
8.1. Mastering-Overcoming Size and Laser Power Limitations
8.1. Mastering-Overcoming Size and Laser Power Limitations
A major bottleneck for applications using large size vHOEs has been the master hologram
A major
recording bottleneck
which, if recordedfor applications
in traditionalusing large size
full aperture vHOEs
mode, needshasexpensive,
been the master
large and hologram
high
recording which, if recorded in traditional full aperture mode, needs
precision optical equipment and high power laser with long coherence length to achieve theexpensive, large and high
precision optical equipment and high power laser with long coherence length to
minimum needed intensities in the plane of the recording medium. Furthermore, the recording setup,achieve the minimum
needed intensities
especially inpath
the object the plane
needsof tothe recording medium. Furthermore, the recording setup, especially
be rearranged.
the object
A largepath needsvHOE
master to be rearranged.
can be recorded alternatively using only standard size and off the shelf
A large master vHOE
optics and low power lasers, canifbethey
recorded alternatively
are combined withusing
an x,only standard stage
y-translation size and off the
[58,59]. shelfsub-
Small optics
and low power
holograms lasers,
(Hogels) if they
are are combined
recorded and stitched an x,toy-translation
withnext each other to stage [58,59].
generate theSmall
largesub-holograms
vHOE. The
(Hogels) are recorded and stitched next to each other to generate the large
Hogels can be recorded by using a physical object generating the object beam or which is much vHOE. The Hogels
morecan
flexible by a spatial light modulator (amplitude or phase SLM) in the object beam. The usage of aby
be recorded by using a physical object generating the object beam or which is much more flexible
programmable SLM gives us a very high flexibility to generate various types of vHOEs in the same
optical and mechanical setup. In the following we review a holographic printing system and one
example of a printed vHOE in Bayfol® HX film, both given in [57].
Polymers 2017, 9, 472 26 of 35

a spatial light modulator (amplitude or phase SLM) in the object beam. The usage of a programmable
SLM gives us a very high flexibility to generate various types of vHOEs in the same optical and
mechanical setup. In the following we review a holographic printing system and one example of
aPolymers
printed vHOE ®
2017, 9, 472in Bayfol HX film, both given in [57]. 26 of 35

8.2.
8.2. Experimental
Experimental Setup
Setup of
of the
the HOE
HOEPrinter
Printer
The
The holographic
holographic printing
printing system
system described
described here
here records
records functional
functional transmission
transmission type
type vHOEs.
vHOEs.
The x, y translation stage that we used allows the recording of vHOE sizes up to 600 mm×× 600
The x, y translation stage that we used allows the recording of vHOE sizes up to 600 mm 600 mm.
mm.
Figure
Figure1818shows
showsthetheoptical
opticaland
andmechanical
mechanicallayout
layoutofofthe
theprinting
printingsystem.
system.A Aphase
phaseonly
onlySLM
SLMisisused
used
to
toform
formthe
theobject
objectbeam
beamofofaaHogel.
Hogel.More
Moredetails
detailson
onthe
thecomponents
componentsusedusedare
aregiven
givenin
in[57].
[57].

Figure 18.
Figure Setupof
18. Setup ofthe
theholographic
holographicprinter
printer in
inwhich
whichaa600
600mm
mm×
× 600
600 mm
mm large
large CGH can be
CGH can be exposed
exposed [57].
[57].

8.3. Application
8.3. Application Example:
Example: Off-Axis
Off-Axis vHOE
vHOE Lens
Lens
Oneexample
One examplefor
foraafunctional
functionalvHOE
vHOEisisaaholographic
holographicoff-axis
off-axistransmission
transmissionlens.
lens.

8.3.1. Recording
8.3.1. Recording Scheme
Scheme
Figure 19
Figure 19 (left)
(left) shows
showsthe thefull
fullaperture
aperturerecording
recordingsetupsetupforforaaholographic
holographicoff-axis
off-axistransmission
transmission
lens. In this
lens. this case,
case, the
the object
object toto record
record is
is realized
realized byby aa very
very small
small circular
circular aperture
aperture (pin
(pin hole)
hole) at
at the
the
location of
location of the
the desired
desired focal
focal point
pointand
andaaplane
planereference
referencewavewaveisisused.
used. Using
Using the
thephase
phaseconjugated
conjugated
referencebeam
reference beamfor forreplay
replaywillwillreconstruct
reconstructthe thereal
realimage
imageof ofthis
thisvHOE,
vHOE,whichwhichisisthe
thefocal
focalpoint
pointofofthis
this
off-axislens,
off-axis lens,asasshown
shownin inFigure
Figure19 19(right).
(right). For
For vHOEs
vHOEs of of the
the size
size of
of aa few
few centimeters
centimeters this
this recording
recording
process isis simple
process simple andand easy
easy totosetup.
setup. However
However for for vHOEs
vHOEs of of larger
larger size
size this
this recording
recording process
process will
will
impose immediately
impose immediately severe severe challenges.
challenges. To To form
form the
the large
large plane
plane reference
reference wavewave respective
respective lenses
lenses or
or
collimatingmirrors
collimating mirrorshave havetotobebe at least
at least as large
as large as vHOE.
as the the vHOE. Therefore
Therefore those those components
components will bewill
heavybe
heavy
and and expensive.
expensive. A homogeneous
A homogeneous power distribution
power distribution (no Gaussian)
(no Gaussian) of the recording
of the recording beams in the beams
planein
thethe
of plane
vHOE of isthe vHOE is to
mandatory mandatory
achieve equalto achieve equal
diffraction diffraction
efficiencies efficiencies
across the vHOE.across the vHOE.
Therefore the
Thereforebeams
recording the recording
have to beams have tomuch
be expanded be expanded
wider than much wider than
the vHOE the vHOE
area itself, whicharea itself,
leads to awhich
huge
leads of
waste to power
a hugeand waste
by of power
this a lowand by this
power a low
density or power density times.
long exposure or longOn exposure
the othertimes.
handOn the other
a minimum
hand a minimum power density is needed to overcome the oxygen inhibition (see Section 5). Within
this tradeoff state of the art single-mode, high power lasers limit the practical vHOE size to a few 10
centimeters. Also very stable and actively fringe locked recording setups are needed for long
exposure time.
Polymers 2017, 9, 472 27 of 35

power density is needed to overcome the oxygen inhibition (see Section 5). Within this tradeoff state of
the art single-mode, high power lasers limit the practical vHOE size to a few 10 centimeters. Also very
stable and
Polymers actively
2017, 9, 472 fringe locked recording setups are needed for long exposure time. 27 of 35
Polymers 2017, 9, 472 27 of 35

Figure 19. Analog recording scheme for an off axis transmission lens using a pin hole (left) as an
Figure 19. Analog
Figure 19. Analogrecording
recordingscheme
schemeforfor
an an
off off
axisaxis
transmission lens lens
transmission usingusing
a pinahole
pin(left)
hole as an object;
(left) as an
object.
The phaseThe phase conjugated readout of thetransmission
off axis transmission lens to reconstruct the real image of
object. Theconjugated readoutreadout
phase conjugated of the off
of axis lens to reconstruct
the off axis transmission the real the
lens to reconstruct image
realofimage
the pin
of
the
hole pin hole
(right) (right) [57].
the pin hole[57].
(right) [57].

On the other hand it is also possible to “stitch” such a vHOE of an off axis lens sequentially. The
On the other
otherhandhandititisisalso
alsopossible
possible toto“stitch” such
“stitch” a vHOE
such a vHOE of an
of off
an axis lens lens
off axis sequentially. The
sequentially.
phase function needed in the object beam is generated by starting the design algorithm with a small
phase function needed in the object beam is generated by starting the design
The phase function needed in the object beam is generated by starting the design algorithm with algorithm with a small
circle. On the left side of Figure 20 it is shown how it looks like if each Hogel would be encoded with
acircle.
smallOn the On
circle. left the
sideleft
of side
Figure 20 it is 20
of Figure shown how ithow
it is shown looksit like
looksif like
eachifHogel wouldwould
each Hogel be encoded with
be encoded
the identical phase function of this circle.
the identical
with phasephase
the identical function of this
function ofcircle.
this circle.

Figure 20. (left) A vHOE stitched with identical Hogels. The same function at a different position is
Figure 20.
20. (left)
(left) AA vHOE
vHOE stitched
stitched with
with identical
identical Hogels.
Hogels. The
The same
same function
function at a different
different position
position is
is
Figure
reconstructed. (right) A vHOE stitched with individual Hogels. Each Hogelatisacorrected with respect
reconstructed. (right) A vHOE stitched with individual Hogels. Each Hogel is corrected
reconstructed; (right) A vHOE stitched with individual Hogels. Each Hogel is corrected with respect with respect
to its position on the vHOE, so that a single focal point is reconstructed [57].
to its position on the vHOE, so so that
that aa single
single focal
focal point
point is
is reconstructed
reconstructed [57].
[57].

To reconstruct a single focal point, for each Hogel a deflecting phase function has to be added to
To reconstruct a single focal point, for each Hogel a deflecting phase function has to be added to
To reconstruct
the phase functionaof single focal circle.
the small point, The
for each Hogel has
deflection a deflecting phase
to be chosen function
with hastotothe
respect beposition
added toof
the phase function of the small circle. The deflection has to be chosen with respect to the position of
the
thephase
Hogelfunction of the and
on the vHOE small circle.
the Thefocal
desired deflection
lengthhas(seetoFigure
be chosen with respect to the position of
20 (right)).
the Hogel on the vHOE and the desired focal length (see Figure 20 (right)).
the Hogel on the vHOE and the desired focal length (see Figure 20 (right)).
8.3.2. Experimental Verification
8.3.2. Experimental
8.3.2. Verification
Experimental Verification
As shown in [57] a vHOE comprised of a matrix of 60 × 60 Hogels was recorded. Using low power
As shown
As shown inin [57]
[57] aa vHOE
vHOE comprisedof of amatrix
matrixof of60
60×× 60 Hogels was recorded. Using low power 2
single frequency DPPS laserscomprised
with output apowers of ~100 60
mW,Hogels
total was
power recorded. Using
densities lowmW/cm
of 100 power
single
single frequency DPPS lasers with output powers of ~100 mW, total power densities of 100 mW/cm 22
can befrequency DPPS at
easily achieved lasers withsize
a Hogel output powers
of 520 of ~100
μm × 520 μm. mW, total power
An exposure timedensities
of 200 msofdelivers
100 mW/cm
the total
can be
can be easily
easily achieved
achieved at aa Hogel
Hogelsize
sizeof
of520
520µmμm× × 520 μm.
µm. An
An exposure time
time of 200 ms delivers the
the total
total
dosage of 20 mJ/cm2 2. at
Besides the exposure 520 movement
time the exposure
and settling oftime
200 of
msthe
delivers
x, y-translation
dosage
dosage of 20 mJ/cm
of 20 mJ/cm 2. Besides
. Besides the exposure time the movement and settling time of the x, y-translation
of the x,phase
y-translation
stage contributes most to thethe exposure
total printingtime
timethe
of movement
~25 min. The andindividual
settling timedeflecting function
stage contributes
stage contributes most
most to to the
the total
total printing
printing time
time ofof ~25
~25 min.
min. The
The individual
individual deflecting
deflecting phase
phase function
function
for each Hogel was selected such that a focal distance of 300 mm was reached. A magnified image of
for each Hogel was selected such that a focal distance of 300 mm was reached. A magnified image of
the vHOE plane, taken from its reconstruction direction (see Figure 21) reveals that perfect alignment
the vHOE plane, taken from its reconstruction direction (see Figure 21) reveals that perfect alignment
of the Hogels was achieved and that they cover the complete vHOE area.
of the Hogels was achieved and that they cover the complete vHOE area.
The functionality of the printed off-axis transmission lens was demonstrated by illuminating it
The functionality of the printed off-axis transmission lens was demonstrated by illuminating it
at 30° (On Bragg) with a plane wave with a diameter larger than the size of the vHOE (Figure 22 (left)).
at 30° (On Bragg) with a plane wave with a diameter larger than the size of the vHOE (Figure 22 (left)).
The incident power on the vHOE can be calculated by multiplication of the average incident power
The incident power on the vHOE can be calculated by multiplication of the average incident power
 Polymers 2017, 9, 472 28 of 35

for each Hogel was selected such that a focal distance of 300 mm was reached. A magnified image of
the vHOE plane, taken from its reconstruction direction (see Figure 21) reveals that perfect alignment
of the Hogels was achieved and that they cover the complete vHOE area.
The functionality of the printed off-axis transmission lens was demonstrated by illuminating it at
30◦ (On Bragg) with a plane wave with a diameter larger than the size of the vHOE (Figure 22 (left)).
The incident
Polymers power on the vHOE can be calculated by multiplication of the average incident28power
2017, 9, 472 of 35
density of ~0.7 mW/cm2 with the vHOE size of ~9.7 cm2 to be 6.8 mW. The 1st diffraction order of the
Polymers 2017, 9, 472 28 of 35
vHOEatata adistance
vHOE distanceofof300
300mm
mmisisshown
shownininFigure
Figure2222(right).
(right).The
Thesize
sizeofofthe
thefocal
focalpoint
pointwas
wasmeasured
measured
2 and the power in the focal plane was measured to be 5.3 mW. By this, a real diffraction
toto
bebe~3~3
vHOE mmmm 2 and
at a distancethe of
power in the
300 mm focal plane
is shown was measured
in Figure 22 (right). to
Thebesize
5.3 of
mW. theBy this,
focal a real
point wasdiffraction
measured
efficiencyofof78%
efficiency 78% could
could be
be estimated
estimated including
including Fresnel
Fresnel losses
losses and
and other
other losses.
losses.
to be ~3 mm and the power in the focal plane was measured to be 5.3 mW. By this, a real diffraction
2

efficiency of 78% could be estimated including Fresnel losses and other losses.

Figure 21.21.
Figure Magnified image
Magnified imageofofa a6060× 60 Hogel
× 60 array
Hogel forming
array anan
forming off-axis holographic
off-axis lens.
holographic The
lens. camera
The camera
system
systemwaswasfocused on the
focused image vHOE
on theofvHOE plane and the Hogels are reconstructed “On Bragg” with the
Figure 21. Magnified a 60 × plane andarray
60 Hogel the Hogels
formingare reconstructed
an off-axis “On Bragg”
holographic lens. Thewith the
camera
reconstruction
reconstruction beam [57].
beam [57].
system was focused on the vHOE plane and the Hogels are reconstructed “On Bragg” with the
reconstruction beam [57].

Figure 22. The printed holographic off-axis lens is reconstructed with a plane wave incident under
30° (“On Bragg”) [57]. As the plane wave is larger than the vHOE size a dark square in the 0th diffraction
Figure 22. The printed holographic off-axis lens is reconstructed with a plane wave incident under ◦
Figure
order 22. The
appears. Atprinted
the focalholographic
distance ofoff-axis
300 mmlens
the is
1streconstructed withforms
diffraction order a plane
thewave
focalincident
point. under 30
30° (“On Bragg”) [57]. As the plane wave is larger than the vHOE size a dark square in the 0th diffraction
(“On Bragg”) [57]. As the plane wave is larger than the vHOE size a dark square in the 0th diffraction
order appears.
order appears. At the
the focal
AtVolume distance
distanceof
of300
focalHolographic 300mmmmthe
the1st1stdiffraction
diffractionorder
orderforms
formsthe
thefocal point.
focal point.
8.4. R2R Replication of Optical Elements
8.4. R2R Replication of Volume Holographic Optical Elements
8.4.In [55,56]
R2R we explained
Replication of Volumethe contact copy
Holographic process
Optical for mass replication of vHOEs in a step and
Elements
repeat In process as stated already above and outlined
[55,56] we explained the contact copy process for the mechanical construction
mass replication and optical
of vHOEs beam
in a step and
shaping In [55,56] we explained
of the installed the contact copy process for mass replication of vHOEs in a step and repeat
repeat process as stated replication
already above line.and
Figure
outlined 23 displays the scheme
the mechanical of process
construction and steps
opticalfor a
beam
process as stated
transmission vHOE. already
The above andscheme
respective outlined the
for a mechanical
reflection construction
vHOE is shown and
in optical beam shaping of
[55,56].
shaping of the installed replication line. Figure 23 displays the scheme of process steps for a
the Holographic
installed replication line.reduces
replication Figure 23
thedisplays
complexity the scheme of process
of the optical steps
Thefor a transmission vHOE.
transmission vHOE. The respective scheme for a reflection vHOEsetup.
is shown laser
in beam handling
[55,56]. is
The respective
simplified, since scheme
only for
one a reflection
beam needs vHOE
to beis shown
managed. in [55,56].
The master creates an object beam that
Holographic replication reduces the complexity of the optical setup. The laser beam handling is
Holographic
propagates replication reduces the complexity of the optical setup. The laser beam handling
simplified,tosince
the copy
onlyfilm
oneand interferes
beam needswith
to bethemanaged.
illuminationThebeam.
master Properly
createsdesigned,
an objecta master—
beam that
is simplified,
copy process to is since onlyfrom
beneficial one beam needs to
a vibrational be managed.
point ofillumination The master
view allowing creates
for aProperly
more an object
robust beam that
process.
propagates the copy film and interferes with the beam. designed, a master—
copy process is beneficial from a vibrational point of view allowing for a more robust process.
Polymers 2017, 9, 472 29 of 35

propagates to the copy film and interferes with the illumination beam. Properly designed, a master—
copy process
Polymers 2017, 9, is beneficial from a vibrational point of view allowing for a more robust process. 29 of 35
472
Polymers 2017, 9, 472 29 of 35

Figure 23.
Figure
Figure 23. Contact
23. Contact copy
Contact copy scheme
copy scheme
scheme of of aaa transmission
of transmission vHOE. From
transmission vHOE.
vHOE. From left
From left to
left to right:
to right: The
right: The copy
The copy film
copy film is laminated
film is
is laminated
laminated
onto the master
onto the master vHOE;
master vHOE; a laser
vHOE; a laser beam
laser beam illuminates
beam illuminates the master
illuminates the master vHOE “On Bragg”,
master vHOE “On Bragg”, whereby the
Bragg”, whereby the master master
master
vHOE acts
vHOE
vHOE acts as
acts as aaa beam
as beam splitter
beam splitter that
splitter that generates
that generates the
generates the object
the object beam.
object beam. This
beam. This object
object beam
beam interferes
interferes with the
with the
the
residual order
order beam
beam behind
behind the
the master
master vHOE,
vHOE, thereby
thereby creating
creating the
the interference
residual 0th order beam behind the master vHOE, thereby creating the interference pattern that is
0th interference pattern that is
recorded into
recorded
recorded into the
into the copy
copy film;
film; downstream
downstream withinwithin aa few
few seconds
seconds the
the dark
dark reaction
reaction is
is finished;
finished; the the copy
copy
film
film is
is delaminated
delaminated from the master
from the
the master vHOEvHOE and
and protected
protected by
by re-masking
re-masking withwith aa cover
cover film;
film; bleaching
bleaching
and fixing
fixing is
is done
done as
as last
last step
step further
further downstream
downstream
and fixing is done as last step further downstream [55]. [55].
[55].

8.4.1. Roll-To-Roll
8.4.1. Roll-To-Roll (R2R)
Roll-To-Roll (R2R) Mechanics
(R2R) Mechanics
Mechanics
The R2R
The R2R film
film handling
handling is is realized
realized by by aa custom
custom made
made machine
machine that
that was
was designed
designed andand built
built with
with
standard
standard winding
standard winding components
windingcomponents
components according
according to the
according to the author’s
to author’s specifications
specifications
the author’s by
by a machine
specifications a machine builder
builder company.
by a machine builder
company.
Figure
company. Figurethe
24 shows
Figure 24 film
24 shows
shows the film
handling
the film handling
diagram
handling diagram
of diagram of the
the R2R
the R2R machine
of R2R
thatmachine thatwidth
has a web
machine that has aaof
has web width
30 cm.
web of 30
A more
width of 30
cm. A
A more
more
detailed
cm. detailedof
description
detailed description
its individual
description of its
of itsmodules
individual
individual modules
of the of the
theline
replication
modules of replication line
is given in
replication is given
given in
[55,56].
line is in [55,56].
[55,56].

Figure 24.
Figure 24. Film
Film handling
handling diagram
diagram of
of the
the R2R
R2R machine
machine with
with dimensions
dimensions in
in millimeters
millimeters and
and process
process
Figure 24. Film handling diagram of the R2R machine with dimensions in millimeters and process
description [55].
description [55].
[55].
description
The sequence
The sequence of of steps
steps before
before laser
laser exposure
exposure consists
consists ofof unwinding
unwinding thethe copy
copy film,
film, release
release ofof the
the
copy film’s cover foil and lamination of the copy film’s photopolymer surface
copy film’s cover foil and lamination of the copy film’s photopolymer surface on a glass plate on a glass plate
optically bonded
optically bonded withwith the
the master
master hologram.
hologram. Optical
Optical quality
quality lamination
lamination is
is possible
possible onon the
the flat-bed
flat-bed
master plate of 500 mm × 300 mm size due to the inherent tackiness of the free photopolymer
master plate of 500 mm × 300 mm size due to the inherent tackiness of the free photopolymer film film
surface. Moreover, this tackiness also leads to a self-cleaning of the master plate from adhering
surface. Moreover, this tackiness also leads to a self-cleaning of the master plate from adhering dust dust
particles after
particles after aa few
few cycles.
cycles. As
As shown
shown inin Figure
Figure 24,
24, we
we laminate
laminate the
the copy
copy film
film to
to the
the lower
lower side
side of
of the
the
master plate. After lamination the line scan exposure is performed. Then, the
master plate. After lamination the line scan exposure is performed. Then, the copy film is de- copy film is de-
Polymers 2017, 9, 472 30 of 35

The sequence of steps before laser exposure consists of unwinding the copy film, release of the
copy film’s cover foil and lamination of the copy film’s photopolymer surface on a glass plate optically
bonded with the master hologram. Optical quality lamination is possible on the flat-bed master plate of
500 mm × 300 mm size due to the inherent tackiness of the free photopolymer film surface. Moreover,
this tackiness also leads to a self-cleaning of the master plate from adhering dust particles after a few
cycles. 2017,
Polymers As shown
9, 472 in Figure 24, we laminate the copy film to the lower side of the master plate.30After of 35
lamination the line scan exposure is performed. Then, the copy film is de-laminated from the master
in a controlled
laminated from way. For this
the master in athe rubber roll
controlled way.used
Forfor
thislamination
the rubber is moved
roll backwards
used for lamination horizontally
is moved
and the web
backwards tension automatically
horizontally and the web releases
tensionthe copy film directly
automatically releasesat the
the copy
rubber filmroll’s contact
directly at line
the
from the master plate. After that the film is moved forward preparing the next step-and-repeat
rubber roll’s contact line from the master plate. After that the film is moved forward preparing the cycle.
The photopolymer
next step-and-repeatsurface of the
cycle. The copy film—now
photopolymer comprising
surface the film—now
of the copy copied vHOE—iscomprisingre-masked with
the copied
a new cover
vHOE—is foil. Subsequently,
re-masked with a newthe protected
cover copy film is the
foil. Subsequently, moved through
protected thefilm
copy bleaching
is moved station and
through
finally is rewound to a new core.
the bleaching station and finally is rewound to a new core.

8.4.2. The
8.4.2. The Laser
Laser Scanning
Scanning Module
Module
The laser
The laser scanning
scanning module
module moves
moves aa plane
plane wave
wave with
with aa highly
highly elliptical
elliptical intensity
intensity profile
profile at
at aa fixed
fixed
incidence angle
incidence angle across
acrossthe
therecording
recordingplate
plateatat
thethe
time when
time when thethe
copy filmfilm
copy is laminated to thetocopy
is laminated plate.
the copy
The layout is shown in Figure 25.
plate. The layout is shown in Figure 25.

Figure 25. Detailed

Figure Detailedlayout
layoutofofthethe
laser scanning
laser module.
scanning (a) Geometrical
module. model;model;
(a) Geometrical (b,c) sketch
(b,c) of module.
sketch of
The incidence angle θ can be varied between ◦ and 60◦ , whereas the slanted mount of the stage β is
module. The incidence angle θ can be varied 20
between 20° and 60°, whereas the slanted mount of the
fixed β ◦ [55].
tois20fixed
stage to 20° [55].

As
As can
can bebe seen
seen from
from Figure
Figure 25b
25b the
the laser
laser scanning
scanning direction
direction is
is set
set perpendicular
perpendicular to to the
the copy
copy film
film
feed direction. A major benefit is that there is more space available for the scanning
feed direction. A major benefit is that there is more space available for the scanning mechanism and mechanism and
optics
optics to
to shape
shape thethe wave
wave front
front of
of the
the plane
plane wave
wave to to the
the highly
highly elliptical
elliptical intensity profile.
intensity profile.
The
The scanning
scanning module
module deflects
deflects the
the plane
plane wavewaveupwards
upwardsby byaarectangular
rectangularfront frontsurface
surfacemirror.
mirror.
The
The holder of this first front surface mirror is mounted on the optical table that supports the beam
holder of this first front surface mirror is mounted on the optical table that supports the beam
and
and wave
wavefront
frontshaping
shapingoptics. A second
optics. A second front surface
front mirror
surface is mounted
mirror is mountedon a linear translation
on a linear stage,
translation
whereas this linear
stage, whereas this translation stage isstage
linear translation heldisbyhelda custom made frame.
by a custom This frame
made frame. This is mounted
frame on an
is mounted
optical
on an optical table while not being connected to the frame of the R2R machine. The slanted mountthe
table while not being connected to the frame of the R2R machine. The slanted mount of of
linear stage
the linear guarantees
stage guarantees bestbest
balance
balancebetween
between achievable
achievablescan
scanlength
length atata alarge
largerange
rangeof of possible
possible
incidence
incidence angles
angles andand height
height of
of the
the scanning
scanningmodule.
module.To Toadjust
adjustthe
theincidence
incidenceangle anglethethesecond
secondmirror
mirror is
is mounted on a rotational plate, which is connected to a motorized rotary stage.
mounted on a rotational plate, which is connected to a motorized rotary stage. Incidence angles θ of Incidence angles θ
of
thethe scanning
scanning beam
beam towards
towards thethe surface
surface normal
normal of of
thethe master
master between
between 20◦20°
and and6060°
◦ cancan
bebe realized
realized in
in this setup. Therefore, a high flexibility in usable master vHOE designs can
this setup. Therefore, a high flexibility in usable master vHOE designs can be accommodated. Both be accommodated. Both
front
front surface
surface mirrors
mirrors have
have aa size
size of
of130
130mm mm××430 430mmmmand andare
aremade
madeofofN-BK7N-BK7 glass
glasscomprising
comprising a
protected
a protected AgAg coating.
coating.

8.4.3. Optics for Beam Shaping and Wave Front Shaping

The module transforms the circular output laser beam of diameter 2 × wB = 2.25 mm into a plane
wave with a highly elliptical intensity profile with a minor axis of 2 × ws = 40.9 mm and a major axis
of 2 × wl = 954.5 mm (Figure 26). Along the major axis the power is ~70% of the center power if a 400
Polymers 2017, 9, 472 31 of 35

8.4.3. Optics for Beam Shaping and Wave Front Shaping

The module transforms the circular output laser beam of diameter 2 × wB = 2.25 mm into a plane
wave with a highly elliptical intensity profile with a minor axis of 2 × ws = 40.9 mm and a major axis
of 2 × wl = 954.5 mm (Figure 26). Along the major axis the power is ~70% of the center power if
a 400 mm wide aperture is used. In the module the laser beam is first circularly expanded and cleaned
by a spatial filter SF. Second, this circular divergent beam is collimated by a first lens L1 to a 1/e2
diameter d3 = 2 × ws of the short edge of the elliptical intensity profile. Third, a concave cylinder lens
C1 expands the 9,beam
Polymers 2017, 472 along the horizontal direction to a desired length at a certain distance. 31 ofFourth,
35
a second “large” cylinder lens C2 re-collimates the beam along the horizontal axis to a 1/e2 radius
Polymers 2017, 9, 472 31 of 35
“large”
wl that cylinder to
corresponds lens
anC2 re-collimates power
edge-to-center the beam along
ratio the horizontal
of 70% axisedge
along a long to a 1/e radiusllw=l that
of 2length 400 mm.
corresponds
The components
“large” to
cylinderforanlens
edge-to-center
the beam powerthe
expansion
C2 re-collimates ratio
and of 70%
wave
beam along
front
along a long edge
theshaping
horizontal of length
optics
axis were
to a 1/e ll 2=radius
chosen 400 mm.
such The
that all
wl that
components for theedge-to-center
beam expansion and ratio
waveof front shaping optics were of
chosen such that all fitsThe
on
fits on an optical table of 1 m × 4 m size and that the NA of the lenses, especially the cylinder lenses,
corresponds to an power 70% along a long edge length ll = 400 mm.
an optical
components table of
for the1 m × 4 m size
beam complex and
expansionandthat the
andcostly NA
wave lens of the lenses,
frontsurface especially
shapingshapes
optics were the cylinder
chosen lenses, are kept
are kept small (<0.2) to avoid corrected forsuch that allaberrations.
spherical fits on
small
an (<0.2)
optical to of
table avoid
1 m complex
× 4 m sizeand
and costly
that thelens
NA surface
of the shapesespecially
lenses, correctedthe
for spherical
cylinder aberrations.
lenses, are kept
Detailed specifications
Detailed specifications of the optics
of the are
optics arelisted
listedinin[55,56].
[55,56].
small (<0.2) to avoid complex and costly lens surface shapes corrected for spherical aberrations.
Detailed specifications of the optics are listed in [55,56].

Figure 26. Detailed

Figure layout
26. Detailed of of
layout thethe
beam
beamand andwave
wave front shapingoptics.
front shaping optics.
TheThe total
total path
path length
length fromfrom
the the
spatial
spatial filterfilter
(SF) (SF)
to to is
C2 C2about
is about
3 3 m.
m. If Ifthe
theoptional
optional QWP
QWP isis set,
set,the
theHWP
HWP and
andPBS areare
PBS removed [55].[55].
removed
Figure 26. Detailed layout of the beam and wave front shaping optics. The total path length from the
spatial filter (SF) to C2 is about 3 m. If the optional QWP is set, the HWP and PBS are removed [55].
8.4.4. Bleaching Module
8.4.4. Bleaching Module
8.4.4.The
Bleaching Module
bleaching module has to provide a suitable spectrum and intensity (see Section 7) to
The bleaching module has to provide a suitable spectrum and intensity (see Section 7) to
irreversibly
Thebleachbleach the
bleaching residual initiator dyesa and cure residual writing monomers. To avoid7)the
irreversibly the module
residual has to provide
initiator dyes suitable
and curespectrum
residual and intensity
writing (see Section
monomers. to the
To avoid
irreversibly bleach the residual initiator dyes and cure residual writing monomers. To high
damage of the substrate and the recorded photopolymer layer (e.g., deformation by too avoidheat,
the
damage of the by
yellowing substrate
high and the recorded photopolymer layer (e.g., deformation by too high UV heat,
damage of thetoo UVand
substrate dosage, …) an optimum
the recorded result could
photopolymer layer be obtained
(e.g., by a combination
deformation by too highofheat,
yellowing byLED
and VIS too high UV arranged
modules dosage, .like
. . ) anFigure
optimum result could becomponents
obtained by a listed
combination of UV
yellowing by too high UV dosage, …)inan 27. The
optimum details
result of be
could theobtained by a are in [55,56].
combination of UV
and VIS LED modules arranged like in Figure 27. The details of the components are listed
and VIS LED modules arranged like in Figure 27. The details of the components are listed in [55,56]. in [55,56].

Figure 27. Layout of the bleaching module. The intensity of the various light sources at the copy film
plane can be varied by sliding the light sources up and down [55].
Figure 27. Layout of the bleaching module. The intensity of the various light sources at the copy film
Figure 27. Layout of the bleaching module. The intensity of the various light sources at the copy film
plane can be varied by sliding the light sources up and down [55].
8.4.5. can
plane Dosage Response
be varied of the the
by sliding Contact
light Copy Process
sources up and down [55].
8.4.5.In
Dosage
order Response
to compareof the
the Contact
scanningCopy Process
contact copy process with a static exposure, we measured
dosage response curves of vHOE copies recorded
In order to compare the scanning contact copy in process
Bayfol HX
withfilms,
® i.e.,exposure,
a static the relative
we diffraction
measured
efficiency η of the copy as function of the applied dosage D during recording. In a static
dosage response curves of vHOE copies recorded in Bayfol HX® films, i.e., the relative diffractionexposure
this is usually done by varying the exposure time at a constant intensity I 0 during recording. In our
efficiency η of the copy as function of the applied dosage D during recording. In a static exposure
machine
this this can
is usually be implemented
done by varying thebyexposure
using thetime
computer controlled
at a constant shutter.
intensity The impinging
I0 during intensity
recording. In our
for any specific position on the copy recording plate can be measured, e.g., by using a power detector
Polymers 2017, 9, 472 32 of 35

8.4.5. Dosage Response of the Contact Copy Process

In order to compare the scanning contact copy process with a static exposure, we measured
dosage response curves of vHOE copies recorded in Bayfol HX® films, i.e., the relative diffraction
efficiency η of the copy as function of the applied dosage D during recording. In a static exposure
this is usually done by varying the exposure time at a constant intensity I0 during recording. In our
machine this9,can
Polymers 2017, 472 be implemented by using the computer controlled shutter. The impinging intensity 32 of 35
for any specific position on the copy recording plate can be measured, e.g., by using a power detector
head
head with
with aa circular
circular mask
mask having
having aa 11 mm
mm diameter
diameter pinhole
pinhole (which
(which is is beneficial
beneficial due
due toto the
the relatively
relatively
narrow Gaussian envelope along the
narrow Gaussian envelope along the short edge).short edge).
For
For the
the scanning
scanning copy
copy process
process aa dependency
dependency of dosage D
of dosage D on
on the
the scanning speed vv needs
scanning speed needs toto be
be
derived from the machine parameters. Details can be
derived from the machine parameters. Details can be found in [55]. found in [55].
As
As anan example
example we we show
show inin Figure
Figure 28
28 the
the results
results of of the
the copy
copy ofofaadigital
digitalprinted
printed master
master vHOE
vHOE
◦ /plane 30◦ ) that has a diffraction efficiency of 40%.
(plane
(planewave
wave00°/plane wave 30°) 40%. Thus, thethe beam ratio BR
beam ratio BR for
for
the copy is 0.67. We applied scanning speeds
the copy is 0.67. We applied scanning speeds stage v between 2 mm/s and
vstage between 2 mm/s and 55 mm/s 55 mm/s in the scanning case
scanning case
2 in both cases.
and
and used
used0.520.52to to14.19
14.19ssfor
forthe
thestatic exposure.IImeasured
staticexposure. was11.5
measured was 11.5mW/cm²
mW/cmin both cases.

Figure 28.
Figure 28. Evidence
Evidence of of independence
independence on on vibrations
vibrations in
in the
thecontact
contactcopy
copyprocess:
process: using
using the
the scanning
scanning
exposure(circles)
exposure (circles)aasimilar
similardosage
dosageresponse
response curve
curve as as
in in
thethe static
static exposure
exposure (diamonds)
(diamonds) is achieved
is achieved for
various scanning speeds. Thus, vibrations do not deteriorate the diffraction efficiency η as inη the
for various scanning speeds. Thus, vibrations do not deteriorate the diffraction efficiency as in the
copy
copy with spacing case
with spacing case [55]. [55].

As we used the contact copy process the dosage response for the scanning process indeed
As we used the contact copy process the dosage response for the scanning process indeed follows
follows that of the static exposure case. This proves the viability of the presented contact copy process
that of the static exposure case. This proves the viability of the presented contact copy process and its
and its suitability for mass production of vHOEs.
suitability for mass production of vHOEs.
9. Conclusions
9. Conclusions
We describe
We describe the
the Chemistry
Chemistry and and the
the Physics
Physics of of aa new
new class
class of
of holographic
holographic photopolymers
photopolymers
(Bayfol® HX
(Bayfol
® HX film)
film) that
that allow
allow aa pure
pure photonic
photonic processing
processing to to manufacture
manufacture volume
volume holographic
holographic optical
optical
elements (vHOEs). Holographic recording is done by laser interference recording.
elements (vHOEs). Holographic recording is done by laser interference recording. Several variants Several variants of
recording schemes are shown including digital master hologram recording
of recording schemes are shown including digital master hologram recording and mass photonic and mass photonic
replication. The
replication. The photochemistry
photochemistry to to build
build a a light
light sensitive
sensitive material
material isis explained
explained in
in detail
detail and
and the
the
chemical response is based on the reaction driven diffusion process. Consequences
chemical response is based on the reaction driven diffusion process. Consequences on the recording on the recording
conditionsare
conditions areexplained.
explained.
vHOEs in photopolymers are
vHOEs in photopolymers are aa key
key enabler
enabler totomany
manynew newapplications.
applications. While
While vHOEs
vHOEs do do not
not
represent a standalone product and need to work in conjunction with other components,
represent a standalone product and need to work in conjunction with other components, including including
light source,
light source, spatial
spatial light
light modulators
modulators andand alike,
alike, they
they need
need to to be
be customized
customized towards
towards specific
specific optical
optical
properties. This film media technology is an enabler towards those needs and links itself to classic
film processing. Major consumer applications are in development like augmented reality glasses and
head-up displays that will change the way we interact with the world in the future. The combination
of Physics and Chemistry has been the driver for the development of these materials, which now will
bridge towards consumer electronics, the automotive industry and beyond.
Polymers 2017, 9, 472 33 of 35

properties. This film media technology is an enabler towards those needs and links itself to classic
film processing. Major consumer applications are in development like augmented reality glasses and
head-up displays that will change the way we interact with the world in the future. The combination
of Physics and Chemistry has been the driver for the development of these materials, which now will
bridge towards consumer electronics, the automotive industry and beyond.

Acknowledgments: We would like to thank all internal and external contributors to the various
sections of this review and list them in alphabetic order of their family names. Xavier Allonas,
Barthel Bach-Zelewski, Christopher Barner-Kowollik, Carsten Benecke, Francesca Bennet, Horst Berneth,
Francois Deuber, Micheal Gleesson, Fabian Grote, Rainer Hagen, Sven Hansen, Dennis Hönel, Ahmad Ibrahim,
Thomas Junkers, David Jurbergs, Koichi Kawamura, Tim Patrick Kleinschmidt, Eberhard Koch, Christian Ley,
Christel Manecke, Maria Schneider-Baumann, Nicolas Stöckel, Enrico Orselli, Christian Rewitz, John Sheridan,
Andy Volkov, Dominik Voll, Dalibor Vukicevic, Günther Walze, Marc Stephan Weiser, Brita Wewer.
Author Contributions: Thomas Fäcke wrote Section 1 and the Conclusions; Thomas Rölle wrote Section 4;
Friedrich-Karl Bruder wrote Sections 2, 3 and 5–8.
Conflicts of Interest: The authors declare no conflict of interest. The founding sponsors had no role in the design
of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, and in the
decision to publish the results.

References
1. Kogelnik, H. Coupled wave theory for thick hologram gratings. Bell Syst. Tech. J. 1969, 48, 2909–2947.
[CrossRef]
2. Kress, B.C. Restocking the optical designer’s toolbox for next-generation wearable displays (Presentation
Recording). Proc. SPIE 2015, 9579, 957903. [CrossRef]
3. Luminit Transparent Holographic Components for Motorcycle Head-up Display. Available online:
http://www.prnewswire.com/news-releases/sid-announces-2017-display-industry-award-winners-
honorees-reflect-the-state-of-the-art-in-display-technology-300458321.html (accessed on 29 July 2017).
4. HoloPro™. Available online: http://www.holopro.com/en/products.html (accessed on 29 July 2017).
5. Zanutta, A.; Orselli, E.; Fäcke, T.; Bianco, A. Photopolymer based VPHGs: From materials to sky results.
Proc. SPIE 2016, 9912, 99123B. [CrossRef]
6. Hwang, Y.S.; Bruder, F.K.; Fäcke, T.; Kim, S.C.; Walze, G.; Hagen, R. Time-sequential autostereoscopic 3-D
display with a novel directional backlight system based on volume-holographic optical elements. Opt. Express
2014, 22, 9820. [CrossRef] [PubMed]
7. Deng, Q.L.; Su, W.C.; Chen, C.Y.; Lin, B.S.; Ho, H.W. Full color image splitter based on holographic optical
elements for stereogram application. J. Display Technol. 2013, 9, 607–612. [CrossRef]
8. Metamaterial Holography. Available online: http://www.metamaterial.com/metamaterial-holography/
(accessed on 29 July 2017).
9. Holographic Technology Revolutionizes Rear Lighting. Available online: http://www.hella.com/hella-
com/en/Technology-Products-19-10-2016-12601.html (accessed on 29 July 2017).
10. HOPS. Available online: http://www.visionoptics.de/index.php?id=18 (accessed on 29 July 2017).
11. Akbari, H.; Naydenova, I.; Martin, S. Using acrylamide-based photopolymers for fabrication of holographic
optical elements in solar energy applications. Appl. Opt. 2014, 53, 1343–1353. [CrossRef] [PubMed]
12. Castillo, J.E.; Russo, J.M.; Kostuk, R.K.; Rosenberg, G.A. Thermal effects of the extended holographic regions
for holographic planar concentrator. J. Photonics Energy 2011, 1, 015504. [CrossRef]
13. Bruder, F.K.; Hagen, R.; Rölle, T.; Weiser, M.S.; Fäcke, T. From the surface to volume: Concepts for the
next generation of optical–holographic data-storage materials. Angew. Chem. Int. Ed. 2011, 50, 4552–4573.
[CrossRef] [PubMed]
14. Trentler, T.J.; Boyd, E.J.; Colvin, V.L. Epoxy-photopolymer composites: Thick recording media for holographic
data storage. Proc. SPIE 2001, 4296, 259–261. [CrossRef]
15. Trentler, T.J.; Boyd, E.J.; Colvin, V.L. Epoxy Resin−Photopolymer Composites for Volume Holography.
Chem. Mater. 2000, 12, 1431–1438. [CrossRef]
16. Weiser, M.-S.; Bruder, F.K.; Fäcke, T.; Hönel, D.; Jurbergs, D.; Rölle, T. Self-Processing, Diffusion-Based
Photopolymers for Holographic Applications. Macromol. Symp. 2010, 296, 133–137. [CrossRef]
Polymers 2017, 9, 472 34 of 35

17. Bruder, F.K.; Deuber, F.; Fäcke, T.; Hagen, R.; Hönel, D.; Jurbergs, D.; Rölle, T.; Weiser, M.S. Reaction-diffusion
model applied to high resolution Bayfol® HX photopolymer. Proc. SPIE 2010, 7619, 76190I. [CrossRef]
18. Gleeson, M.R.; Sheridan, J.T.; Bruder, F.K.; Rölle, T.; Berneth, H.; Weiser, M.S.; Fäcke, T. Comparison of a new
self-developing photopolymer with AA/PVA based photopolymer utilizing the NPDD model. Opt. Express
2011, 19, 26325–26342. [CrossRef] [PubMed]
19. Rölle, T.; Bruder, F.K.; Fäcke, T.; Weiser, M.S.; Hönel, D.; Stöckel, N. Photopolymerzusammensetzungen für
Optische Elemente und Visuelle Darstellungen. European Patent EP 2 172 505 A1, 7 April 2010.
20. Zhao, G.; Mouroulis, P. Diffusion model of hologram formation in dry photopolymer materials. J. Mod. Opt.
1994, 41, 1929–1939. [CrossRef]
21. Noiret, N.; Meyer, C.; Lougnot, D.J. Photopolymers for holographic recording: V. Self-processing systems
with near infrared sensitivity. Pure Appl. Opt. 1994, 3, 55–71. [CrossRef]
22. Sheridan, J.T.; Lawrence, J.R. Nonlocal-response diffusion model of holographic recording in photopolymer.
J. Opt. Soc. Am. A 2000, 17, 1108–1114. [CrossRef]
23. O’Neill, F.T.; Lawrence, J.R.; Sheridan, J.T. Comparison of holographic photopolymer materials by use of
analytical nonlocal diffusion models. Appl. Opt. 2002, 41, 845–852. [CrossRef] [PubMed]
24. Kelly, J.V.; O’Neill, F.T.; Sheridan, J.T.; Neipp, C.; Gallego, S.; Ortuno, M. Holographic photopolymer
materials: Nonlocal polymerization-driven diffusion under nonideal kinetic conditions. J. Opt. Soc. Am. B
2005, 22, 407–416. [CrossRef]
25. Gallego, S.; Ortuno, M.; Neipp, C.; Fernandez, E.; Belendez, A.; Pascual, I. Improved maximum uniformity
and capacity of multiple holograms recorded in absorbent photopolymers. Opt. Express 2005, 15, 9308–9319.
[CrossRef]
26. Gleeson, M.R.; Sheridan, J.T. A review of the modeling of free-radical photo polymerization in the formation
of holographic gratings. J. Opt. A Pure Appl. Opt. 2009, 11, 1–12. [CrossRef]
27. Sheridan, J.T.; Gleeson, M.R.; Close, C.E.; Kelly, J.V. Optical response of photopolymer materials for
holographic data storage applications. J. Nanosci. Nanotechnol. 2007, 7, 1–11. [CrossRef]
28. Zanutta, A.; Orselli, E.; Fäcke, T.; Bianco, A. Photopolymeric films with highly tunable refractive index
modulation for high precision diffractive optics. Opt. Mater. Express 2016, 6, 252–263. [CrossRef]
29. Drobny, J.G. Radiation Technology for Polymers; CRC: Bocan Raton, FL, USA, 2010; ISBN 978-1-4200-9404-6.
30. Schwalm, R. UV Coatings: Basics, Recent Developments and New Applications; Elsevier: Amsterdam,
The Netherlands, 2007; ISBN 9780444529794.
31. Allonas, X.; Croutxe-Barghorn, C.; Fouassier, J.P.; Lalevée, J.; Malval, J.P.; Morlet-Savary, F. Lasers in
Chemistry: Probing and Influencing Matter; Lackner, M., Ed.; Wiley-VCH: Weinheim, Germany, 2008; Volume II,
pp. 1001–1027. ISBN 978-3-527-31997-8.
32. Allonas, X.; Croutxé-Barghorn, C.; Bögl, K.W.; Helle, N.; GSchreiber, G.A. Ullmann’s Encyclopedia of Industrial
Chemistry; Wiley-VCH: Weinheim, Germany, 2012; ISBN 9783527306732.
33. Dietliker, K. A Compilation of Photo-Initiators Commercially Available for UV Today; SITA Technology Limited:
London, UK, 2002; ISBN 0947798676, 9780947798673.
34. Fouassier, J.P.; Allonas, X.; Lalevée, J.; Dietlin, C. Photochemistry and Photophysics of Polymer Materials;
Allen, N.S., Ed.; Wiley: Hoboken, NJ, USA, 2010; pp. 351–419, ISBN 978-0-470-13796-3.
35. Ibrahim, A.; Allonas, X.; Ley, C.; Kawamura, K.; Berneth, H.; Bruder, F.K.; Fäcke, T.; Hagen, R.; Hönel, D.;
Rölle, T.; et al. High Performance Photoinitiating Systems for Holography Recording: Need for a Full Control
of Primary Processes. Chem. A Eur. J. 2014, 20, 15102–15107. [CrossRef] [PubMed]
36. Ley, C.; Carré, C.; Ibrahim, A.; Allonas, X. Holographic Materials and Systems: Application of High Performance
Photoinitiating Systems for Holographic Grating Recording; Naydenova, I., Ed.; InTech: Rijeka, Croatia, 2017.
[CrossRef]
37. Schilling, M.L.; Colvin, V.L.; Dhar, L.; Harris, A.L.; Schilling, F.C.; Katz, H.E.; Wysocki, T.; Hale, A.; Blyler, L.L.;
Boyd, C. Acrylate oligomer-based photopolymers for optical storage applications. Chem. Mater. 1999, 11,
247–254. [CrossRef]
38. Davidenko, N.; Garcia, O.; Sastre, R. The efficiency of titanocene as photo-initiator in the polymerization of
dental formulations. J. Biomater. Sci. Polym. Ed. 2003, 14, 733–746. [CrossRef] [PubMed]
39. Fouassier, J.P.; Morlet-Savary, F. Photopolymers for laser imaging and holographic recording: Design and
reactivity of photosensitizers. Opt. Eng. 1996, 35, 304–312. [CrossRef]
Polymers 2017, 9, 472 35 of 35

40. Adair, P.C.; Gottschalk, P. Photo-Initiators and Photosensitive Compositions Containing Such Photo-Initiators.
European Patent EP 0 368 629 A2, 8 November 1988.
41. Davies, A.G.; Roberts, B.P. Bimolecular homolytic substitution by tert-butoxy radicals at metal atoms.
Organomet. Chem. 1969, 19, 17–18. [CrossRef]
42. Murphy, S.T.; Zou, C.; Miers, J.B.; Ballew, R.M.; Dlott, D.D.; Schuster, G.B. Tetraarylborates {[Ar]4B-}:
Estimation of oxidation potentials and reorganization energies from electron-transfer rates. J. Phys. Chem.
1993, 97, 13152–13157. [CrossRef]
43. Ermoshkin, A.A.; Nikolaeva, E.S.; Neckers, D.C.; Fedorov, A.V. New tetraalkylborate initiators for remote
polymerization. Macromolecules 2008, 41, 9063–9066. [CrossRef]
44. Polykarpov, A.Y.; Hassoon, S.; Neckers, D.C. Tetramethylammonium tetraorganylborates as coinitiators
with 5,7-diiodo-3-butoxy-6-fluorone in visible light polymerization of acrylates. Macromolecules 1996, 29,
8274–8276. [CrossRef]
45. Rehm, D.; Weller, A. Kinetics of fluorescence quenching by electron and H-atom transfer. Isr. J. Chem. 1970, 8,
259–271. [CrossRef]
46. Neumann, M.G.; Pastre, I.A.; Previtali, C.M. Comparison of photoinduced electron transfer to singlet and
triplet states of safranine T. J. Photochem. Photobiol. A 1991, 61, 91–98. [CrossRef]
47. Jakubiak, J.; Allonas, X.; Fouassier, J.P.; Sionkowska, A.; Andrzejewska, E.; Linden, L.A.; Rabek, J.F.
Camphorquinone–amines photo-initating systems for the initiation of free radical polymerization. Polymer
2003, 44, 5219–5226. [CrossRef]
48. Dossot, M.; Allonas, X.; Jacques, P. Unexpected correlation between electronic coupling and excited state
redox properties in PET. Phys. Chem. Chem. Phys. 2002, 4, 2989–2993. [CrossRef]
49. Lambert, C.R.; Kochevar, I.E. Electron transfer quenching of the Rose Bengal triplet state. Photochem. Photobiol.
1997, 66, 15–25. [CrossRef] [PubMed]
50. Allonas, X.; Jacques, P. Factors affecting adiabaticity in bimolecular photoinduced electron transfer reaction
between anthracene derivatives and organic donors. Chem. Phys. 1997, 215, 371–378. [CrossRef]
51. Berneth, H.; Bruder, F.K.; Fäcke, T.; Hagen, R.; Hönel, H.; Jurbergs, D.; Rölle, T.; Weiser, M.S. Holographic
recording aspects of high-resolution Bayfol® HX photopolymer. Proc. SPIE 2011, 7957, 79570H. [CrossRef]
52. Barner-Kowollik, C.; Bennet, F.; Schneider-Baumann, M.; Voll, D.; Rölle, T.; Fäcke, T.; Weiser, M.S.;
Bruder, F.K.; Junkers, T. Detailed investigation of the propagation rate of urethane acrylates. Polym. Chem.
2010, 1, 470–479. [CrossRef]
53. Weiser, M.S.; Rölle, T.; Bruder, F.K.; Fäcke, T.; Hönel, D. Method for Producing a Holographic Film. U.S. Patent
#14/284,488, 3 November 2009.
54. Weiser, M.S.; Rölle, T.; Bruder, F.K.; Fäcke, T.; Hönel, D. Photopolymer Formulations Having the Adjustable
Mechanical Modulus GUV . U.S. Patent #14/543,963, 3 November 2009.
55. Bruder, F.K.; Fäcke, T.; Grote, F.; Hagen, R.; Hönel, D.; Koch, E.; Rewitz, C.; Walze, G.; Wewer, B. Mass
production of volume holographic optical elements (vHOEs) using Bayfol® HX photopolymer film in
a roll-to-roll copy process. Proc. SPIE 2017, 10127, 101270A. [CrossRef]
56. Bruder, F.K.; Fäcke, T.; Grote, F.; Hagen, R.; Hönel, D.; Koch, E.; Rewitz, C.; Walze, G.; Wewer, B. Performance
optimization in mass production of volume holographic optical elements (vHOEs) using Bayfol® HX
photopolymer film. Proc. SPIE 2017, 10233, 102330G. [CrossRef]
57. Bruder, F.K.; Fäcke, T.; Hagen, R.; Hönel, D.; Kleinschmidt, T.P.; Orselli, E.; Rewitz, C.; Rölle, T.; Walze, G.
Diffractive optics in large sizes: Computer-generated holograms (CGH) based on Bayfol® HX photopolymer.
Proc. SPIE 2015, 93850, 93850C. [CrossRef]
58. Yamaguchi, M.; Koyama, T.; Endoh, H.; Ohyama, N.; Takahashi, S.; Iwata, F. Development of a prototype
full-parallax holoprinter. Proc. SPIE 1995, 2406, 50. [CrossRef]
59. Shirakura, A.; Kihara, N.; Baba, S. Instant holographic portrait printing system. Proc. SPIE 1998, 3293, 246.
[CrossRef]

© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).