Short communication
Received: 29 November 2013 Revised: 23 May 2014 Accepted: 18 June 2014 Published online in Wiley Online Library
(wileyonlinelibrary.com) DOI 10.1002/jrs.4550Depth profiling laminated glass with a fiber
optic probe customized for adjustable working
distance†
Odile Madden,a* Gary Gordon,b Kim Cullen Cobba and Alex M. SpencercFiber optic probes allow for in situ characterization of cultural heritage objects and analysis of materials that are difficult to
access. Positioning these probes is challenging in terms of focal distance, angle of analysis, and stability. Modifications to im-
prove control include stabilizing the probe against a stationary surface, typically mediated by a tripod, or against the artifact
itself with a distance regulating sheath that fixes the focal point at the object surface. The first makes the system less portable,
while the second eliminates depth profiling capability. An adjustable working distance adapter was created that allows the op-
erator to position a fiber optic probe against the surface of a transparent artifact and move the working distance up to 6mm into
thematerial while excluding ambient light. The hollow adapter contains no opticalfiber, lenses, orwindows, so optics are dictated
by the fiber optic probe. The tool was created to study the polymeric interlayers in laminated safety glass used in early 20th
century aviation and also could be applied to contemporary laminated glass, othermultilayer transparent objects, and substances
in transparent containers. Published 2014. This article is a U.S. Government work and is in the public domain in the USA.
Keywords: portable Raman; adjustable working distance; laminated glass; fiber optic; cellulose nitrate* Correspondence to: Odile Madden, Museum Conservation Institute,
Smithsonian Institution, 4210 Silver Hill Road, Suitland, MD 20746, USA. E-mail:
maddeno@si.edu
† This article is part of the special issue of the Journal of Raman Spectroscopy en-
titled “Raman in Art and Archaeology 2013” edited by Polonca Ropret and Juan
Manuel Madariaga.
a Museum Conservation Institute, Smithsonian Institution, 4210 Silver Hill Road,
Suitland, MD, 20746, USA
b National Air and Space Museum, Steven F. Udvar-Hazy Center, Smithsonian
Institution, Chantilly, VA, 20151, USA
c National Air and Space Museum, Smithsonian Institution, Washington, DC,
20560, USAIntroduction
Fiber optic probes are valuable tools for in situ analysis of cultural
heritage objects. They allow for analysis of materials that are not
easily accessible because of their location, orientation, fragility, or
sampling restrictions.[1] Positioning these probes is challenging in
terms of focus, angle of analysis, and stability.[2–4] The benefits of
a lightweight probe quickly are negated by a shaky human hand.
This is particularly true for longer analyses that often are required
when the Raman signal is weak or noisy, due to either the nature
of the sample or the operator’s desire to reduce risk of damage to
the artifact by limiting the excitation laser power.
Control can be improved by stabilizing the probe against a sta-
tionary surface, typically mediated by a tripod or other stand, or
against the artifact itself with a sheath that fits over the end of
the fiber optic probe and fixes the focal point at the object sur-
face.[2–5] Tripod type solutions now are used routinely for analysis
of vertical surfaces including paintings and architectural ele-
ments and can be equipped with adjustable fine positioning
stages for focusing. However, tripods and other stands have lim-
itations. First, the angle of analysis usually is restricted and may
not suit the orientation of the artifact. Bulky stands are cumber-
some, can create tripping hazards, and are liable to tip into arti-
facts; this is particularly worrisome in awkward gallery spaces.
Furthermore, interference from ambient light can prevent collec-
tion of usable spectra, but working in the dark (i.e., outdoors or in
a public space) often is not possible. When darkness is an option,
risk to the artifact is increased because the operators cannot see
what they are doing. Excluding ambient light by covering the an-
alytical set up with draped fabrics or aluminum foil is clumsy and
tiring for the team, which also increases risk to artifacts.[4]
Another option is a detachable distance regulating tube that
fits over the fiber optic probe and is placed directly against theJ. Raman Spectrosc. (2014) Published 2014. This aartifact surface.[2,5] Part of the allure of Raman spectroscopy is
non-contact analysis, which can be important for fragile artifact
surfaces. However, with cultural heritage the overarching goal is
to avoid damage (or undue risk to) the artifact. Working together,
a conservator and scientist may decide that stabilizing the instru-
ment against the artifact is a valid option. There are artifacts that
will not be disturbed by direct, gentle contact with the probe
head, and this is true of the chemically stable window glass de-
scribed in this communication. A fiber optic probe sheathed with
a distance regulating tube is lightweight, which can be safer for
fragile artifacts than a handheld spectrometer, the typical weight
of which can range from 0.5–6 lbs (0.3–2.7 kg). It is easy to posi-
tion perpendicular to any exposed surface, and the tube excludes
ambient light. The drawback is that the ability to adjust the depth
of analysis is lost.
As part of a study of laminated safety glass used in early avia-
tion, the authors designed a tube adapter that fits over the fiber
optic probe head and allows the operator to adjust the workingrticle is a U.S. Government work and is in the public domain in the USA.
O. Madden et al.distance while stabilizing the probe against the artifact surface
and eliminating ambient light. This allowed for analysis of poly-
meric interlayers in historic laminated safety glass.Adjustable focal length adapter
The adapter is constructed of two machined aluminum tubes that
are threaded one over the other (Fig. 1a and b). The adapter’s inner
tube fits over the fiber optic probe shaft. Its inside diameter is
matched to outside diameter of the probe shaft, and they have the
same length. A rubber O-ring in a concentric groove on the tube in-
terior provides a friction fit. The second aluminum tube is shorter,
and its inside diameter is matched to the outside diameter of the first
tube, over which it is threaded. The far end of the second tube is fin-
ished flat and is the contact point for the artifact. The adapter is hol-
low, without optical fiber, lenses, or windows, so the focusing
parameters and confocality are dictated by the fiber optic probe
and the spectrometer.
To collect a spectrum, the adapted probe is placed against the
material that will be analyzed, and the depth of the focal point is
adjusted by winding the outer tube back or forth. Maximum focal
depth is achieved when the ends of the two tubes are even and is
limited to the focal distance of the fiber optic probe (Fig. 1b and
c). Winding the outer tube forward moves the focal point back
out toward the surface of the material.
For our prototype, working distance is continuously adjustable in
micron scale increments, where one full revolution changes the focal
length by 1.19mm (3/64 in.). The entire focal range of our fiber optic
probe (5.9mm, ~1/4 in.) is spanned in 4.96 revolutions. The travelFigure 1. Adjustable working distance adapter. (a) Cross section view of ada
to analyze an object’s surface (left) and maximum depth (right), and (c) schem
wileyonlinelibrary.com/journal/jrs Published 2014. This a
and is in the puwas chosen empirically but could be calculated precisely to suit the
lens of any focused fiber optic probe and desired analytical depth.Case study: laminated safety glass
The idea for an adjustable working distance adapter was con-
ceived during a technical study of polymers used in early aviation
by the Smithsonian’s Museum Conservation Institute and National
Air and Space Museum. A portable B&W Tek MiniRam II dispersive
Raman spectrometer with a fiber optic probe (Model #BAC100-
785) was used to analyze nine aviator goggles with lenses of lam-
inated safety glass, where the polymer of interest is encased be-
tween two sheets of glass and could not be accessed directly.
This was a new approach to studying laminated glass. Previous
published studies, outside the cultural heritage field, characterize
the polymer or the polymer-glass interface using experimentally
prepared, open face mock-ups or after deconstructing the lami-
nate.[6–8] Neither is appropriate for studying intact historic artifacts.
Any characterization of the polymer interlayer would have to be
performed through the glass, so Raman spectroscopy was a valu-
able tool.
Goggles that survive from the World War I period are fragile
and do not lie in an orientation that suits typical analytical setups
that use a tripod or stand. Other historic aviation artifacts studied,
such as airplane windshields, also are challenging to access with
a spectrometer because of their height above the ground and
recessed position relative to the airplane’s edge. A fiber optic
probe modified with the adjustable working distance adapter
has been a versatile solution in both cases.[9]pter mounted on fiber optic probe, (b) photographs of adapter configured
atic of how focal positions in (b) would apply to analysis of laminated glass.
rticle is a U.S. Government work
blic domain in the USA.
J. Raman Spectrosc. (2014)
Figure 2. Raman spectra of aviator goggles in National Air and Space Museum collection. Upper spectral group shows two spectra collected at differ-
ent depths into proper left lens of A19620048000, with lower spectrum representing the front glass pane and upper spectrum representing the polymer
interlayer. Lower spectral group shows polymer interlayer of four goggle lenses (from top: A19620048000, A19500122000, A19790846000,
A19761327000). Whereas upper three spectra are consistent with a World War I era formulation of cellulose nitrate plasticized with camphor, the
lowermost spectrum suggests those goggles were manufactured later.
Adjustable working distance adapter for depth profiling laminated glassFor this study, spectra were co-additions of 64 one-second
scans collected at 785 nm excitation and 10 cm
1 resolution.
The 1.5m fiber optic assembly combines one excitation and
one collection fiber into a single tubular probe that both focuses
the excitation beam and collects the scattered light through a flat
quartz window with antireflective coating, 5.9mm working
distance, and 85μm focused analytical spot. The upper spectral
group in Figure 2 shows two spectra collected at different posi-
tions in a lens from one set of goggles. The lower spectrum is of
the glass pane with a broad spectral feature at 1000–2000 cm
1.
Although window glass is a relatively weak scatterer, its signature
dominates both spectra because of the thickness of the glass
panes relative to the polymer interlayer in these goggles. The
upper spectrum shows peaks related to the interlayer; the most
prominent of which are a sharp band at 650 cm
1 peak character-
istic of the plasticizer camphor and a weaker, less defined band at
850 cm
1 that is consistent with cellulose nitrate but inconsistent
with cellulose acetate (a less likely but plausible polymer from that
era). There may be some contribution from camphor at 855 and
863 cm
1 as well.[9–11] Even though the depth resolution of the
spectrometer and probe is too coarse to isolate the sub-millimeter
thick polymer layer, the ability to finely position the focal point at
the center of that layer allows its signal to be maximized relative to
the surrounding glass. A second set of spectra illustrate that this
formulation of cellulose nitrate plasticized with camphor was
typical of laminated safety glass of the World War I period and
was detected in most of the goggles with laminated lenses. An
exception is Accession #A19761327000, the spectrum of which
suggests a later fabrication date.J. Raman Spectrosc. (2014) Published 2014. This article is a U.S
and is in the public domainDiscussion and conclusion
Improving interfaces between Raman spectrometers, artifacts,
and operators is a central challenge for successful in situ analysis
of cultural heritage objects. Transportable Raman spectrometers
are a potential boon, but true portability and usefulness in a
practical setting can be elusive. Handheld spectrometers are
lightweight to carry but can become cumbersome and even
unsafe for long analyses of fragile, awkwardly oriented objects.
Tripods offer stability at some cost of portability and risk to
the artifact. A fiber optic probe is lightweight, offers greater
freedom of movement, and can be positioned in any orienta-
tion. For artifacts that are robust to direct contact, we present
a hollow, adjustable working distance adapter that converts a
fiber optic probe into a depth profiling device that is stationary
relative to the artifact and excludes ambient light. The design is
elegantly simple. The adapter contains no lenses, so optical
quality and confocality are dictated by the fiber optic probe
and spectrometer. The adapter’s threaded aluminum tubes are
finely machined, which allows for fine positioning of the focal
point to maximize the Raman signal from thin layers. In addition
to laminated glass, potential applications include gemstones,
reverse glass paintings, face mounted photographs, and sub-
stances in transparent containers.
Acknowledgements
The authors are grateful to Judith Cherwinka for purchase of the
B&W Tek spectrometer, Don Williams for fabricating the adapter’s. Government work
in the USA.
wileyonlinelibrary.com/journal/jrs
O. Madden et al.first iteration, and Lauren Horelick. This research was funded by
the National Park Service and the National Center for Preserva-
tion Technology and Training (Grant #MT-2210-10-NC-10). Its
contents are solely the authors’ responsibility.
References
[1] P. Vandenabeele, F. Verpoort, L. Moens, J. Raman Spectrosc. 2001, 32, 263.
[2] D. Lauwers, A. G. Hutado, V. Tanevska, L. Moens, D. Bersani, P.
Vandenabeele, Spectrochim. Acta A. 2014, 118, 294.
[3] P. Vandenabeele, K. Castro, M. Hargreaves, L. Moens, J. M. Madariaga, H.
G. M. Edwards, Anal. Chim. Acta 2007, 588, 108.wileyonlinelibrary.com/journal/jrs Published 2014. This a
and is in the pu[4] P. Vandenabeele, J. Tate, L. Moens, Anal. Bioanal. Chem. 2007, 387,
813.
[5] Detachable distance regulator from B&W Tek: http://bwtek.com/
products/bac100/ (accessed November 24, 2013).
[6] D. J. David, T. N. Wittberg, J. Adhesion 1984, 17, 231.
[7] N. M. S. El-Din, M. W. Sabaa, Polym. Degrad. Stab. 1995, 47,
283.
[8] M. Kawagoe, Y. Doi, N. Fuwa, T. Yasuda, K. Takata, J. Mater. Sci. 2001
36, 5161.
[9] O. Madden, K. Cullen Cobb, A. M. Spencer, J. Raman Spectrosc. 2014,
45, 1215.
[10] C. Paris, C. Coupry, J. Raman Spectrosc. 2005, 36, 77.
[11] H. G. M. Edwards, Anal. Bioanal. Chem. 2010, 397, 2677.rticle is a U.S. Government work
blic domain in the USA.
J. Raman Spectrosc. (2014)