Detection of fluorine in skibases and skiwaxes - Gliding
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Gliding 3(2021)13-19
Detection of fluorine in skibases and
skiwaxes
Matthias Scherge, Team Snowstorm, 76316 Malsch
Abstract
The Fluorine Tracker (FT) is a novel measuring device for the detection of flu-
orine in ski waxes and ski soles, which is based on Selective Broadband Re-
flection and Absorption Spectroscopy (SBRAS) and realizes the surface anal-
ysis with visible light, infrared and ultraviolet radiation. The instrument can
detect fluorine concentrations in the single-digit atomic percent range. Fur-
thermore, the method allows measurements at different depths of wax or ski
sole, from a few nanometers to several 100 micrometers.
Introduction tion prohibition. He stated that the
Per- and polyfluoroalkyl substances only way to develop a practical test
(PFAS) comprise several thousand flu- to be used for FIS competitions will
orinated compounds with the unique be to ban all fluor products and not
tribological property of water repel- only the specific molecules which are
lency, thereby reducing the negative banned from production. The Coun-
effect of capillary attraction which cil thereafter decided that the use
slows down the ski [1, 2]. Because of fluorinated ski waxes, which have
of the strong carbon-fluorine bonds been shown to have a negative en-
the degradation of PFAS in nature is vironmental and health impact were
highly retarded and some substances banned for all FIS disciplines from the
persist indefinitely [3]. Some PFAS 2020/2021 season” [6].
are toxic and have been linked to
negative health impacts. The envi- State of the art
ronmental and health impacts of flu- Methods based on different physical
orinated ski waxes have also been principles exist for the detection of
scientifically confirmed [4, 5]. Regu- fluorine. These methods are partly
lations have been established in the quantitative and obtain the informa-
USA that are already in place and tion from different depths of the ma-
were implemented in the EU from July terial (information depth). What all
2020. methods have in common is that their
At the FIS Council Meeting, 23rd use in competitions is only possible
November 2019, in Konstanz “Coun- to a very limited extent. The restric-
cil Member Martti Uusitalo as chair- tion results from the construction and
man of Vauhti ski wax provided a the way of handling of the device.
detailed insight into the current sit- As a rule, the devices are designed
uation concerning the specific ele- for broad laboratory use. In terms
ments which are subject to produc- of price, the devices range between
20 T€ and more than 1 million €.
Open Access Journal – ISSN 2509-9442 – snowstorm-gliding.de 13
Figure 1: Left: Fluori-
nated ski base. Right:
Ski base after removal
of material with a
thickness of 100 nm.
X-Ray Photoelectron Spectroscopy results down to concentrations of less
(XPS) than 1 atomic percent.
XPS can be used to determine the
concentration of fluorine in wax and As an example the analysis of a
ski base as well as the penetration of fluorinated ski base is presented in
wax into the ski base. The surface Fig. 1. With 550 counts the fluo-
of the base is irradiated with an x- rine F1s core level peak was clearly
ray beam, causing binding electrons identified. When by a material re-
to leave the surface. Due to the very moval procedure (sputtering with Ar-
precise measurement of the electron gon ions) 100 nm were stripped off,
energy, i.e., the binding energy, the the signal dropped to 150 counts.
atom that emitted this electron can When it is assumed that fluorine
be identified. XPS is highly surface- is equally distributed inside the ski
sensitive with an information depth of base, then fluorine must have mi-
less than 3 nm. In addition, after cal- grated from the inside to the surface,
ibration using an inbuilt gold sample, causing the higher concentration at
XPS is quantitative and gives reliable the very surface.
Figure 2: Absorbance
FTIR spectrum of a
waxed ski base.
FTIR-ATR Spectroscopy (FTIR-ATR) method. FTIR means that
Another measurement method for the raw signal, i.e., the light inten-
the detection of fluorine on the sity, is not measured as a function
ski base is infrared spectroscopy of wavelength, but as a function of
and here especially the Fourier- the position of a mirror inside the in-
Transformed Infrared Spectroscopy terferometer. Therefore, this signal
Open Access Journal – ISSN 2509-9442 – snowstorm-gliding.de 14
must first be Fourier-transformed to able to detect trace amounts, but re-
obtain the usual IR representation of quires a certain number of wax parti-
light intensity as a function of energy cles. The use of this method for com-
in terms of wavenumbers, see Fig. 2. petition is very limited. Wax samples
ATR means attenuated total reflec- taken from a ski have to be sent to a
tion. This principle has become the laboratory for analysis.
standard technique for measuring FT-
IR spectra. Here, infrared light passes Thermogravimetric analysis (TGA)
through a crystal of an IR-transparent and differential scanning calorimetry
material (diamond, ZnSe, or germa- (DSC)
nium) and then interacts with a sam- With TGA and DSC further methods
ple that has previously been pressed exist that allow conclusions to be
onto the crystal. The resulting spec- drawn about the composition of ski
trum contains information about all waxes.
substance-specific properties. In TGA, the change in mass of
One of the first IR measurements the wax sample is measured as a
to identify the C-F valence vibra- function of temperature. The sample
tion in waxes is described in [7] and is heated in an inert, temperature-
proved the presence of the symmet- stable crucible in an oven. A mi-
ric and an asymmetric valence vibra- crobalance is used to record the mass
tion at 1050 and 1300 cm−1 . change. A loss of mass may re-
The detectable species are molec- sult from a decomposition reaction or
ular groups, the detection limit is ap- from a volatile component evaporat-
prox. 0.1 - 1% and the information ing. An increase in mass can result
depth is 1 to 10 μm, depending on from chemical reactions, such as oxi-
the angle of impinging light. dation. The method can be used, for
In order to keep the device func- example, to detect whether the fluo-
tioning, the ATR crystal has to be rine component is present as a di- or
cleaned regularly. The measuring tri-block copolymer.
time per spot is between 30 s and 1 In DSC, sample and reference
minute. For interpretation, the spec- sample are heated simultaneously so
tra have to be normalized using for that both crucibles always have the
instance the polyethylene peak. The same temperature. The two crucibles
method is not quantitative. are subjected to a temperature-time
program and the heat flow difference
X-ray fluorescence analysis (XRF) between the two samples is recorded.
A third method, which is currently In the spectrum, physical and chem-
used for the field detection of flu- ical reactions can be read off in the
orine, is X-ray fluorescence analy- form of endothermic or exothermic
sis. XRF uses the effect of fluores- peaks. By comparing the peak ar-
cence following excitation with an X- eas, a statement can be made about
ray. Electrons close to the nucleus the proportion of the fluorine compo-
are freed, which then leave the sur- nent. Both methods are not suitable
face and are detected by a radiation for field use.
detector. The samples are not de-
stroyed. The detection limit of XRF Gas chromatography (GC)
is about one microgram per gram GC is an analytical method for sepa-
(ppm). The measurement takes place rating mixtures into individual chem-
in a vacuum chamber. It enables ical compounds. GC is only applica-
the identification and concentration ble to components that are vapor-
determination of elements with an izable in gaseous or undecomposed
atomic number of 5 or more. The form. A gas chromatography-mass
heavier the elements, the better the spectrometry coupling succeeds in
method works. Since fluorine with determining the composition of the
an atomic number of 9 is quite light, vapor space above the sample. This
the detection is sometimes compli- method was used by [9, 10] to an-
cated, so the method had to be opti- alyze 50 commercially available ski
mized for the detection of fluorinated waxes. Perfluorocarbons and per-
waxes [8]. However, XRF is not yet fluoroalkanes compounds were reli-
Open Access Journal – ISSN 2509-9442 – snowstorm-gliding.de 15
ably detected. The measurements centration. The influence of reflectiv-
are time consuming because the gas ity of the ski surface was addressed
species must diffuse through a col- by adapted algorithms combining the
umn to become analyzed. signals of visible (VIS), infrared (IR)
and ultraviolet (UV) light channels. In
How the FT works addition, the impact of colored bases
General function was treated in this way. The device
combines the input of 3 VIS sensors,
The detection of fluorine in ski waxes
2 UV sensors and 4 IR sensors. Both
is based on SBRAS. SBRAS exposes
in UV and IR range along the source-
the ski base to electromagnetic irridi-
sensor axis (inline) and perpendicular
ation with a wide spectral distribu-
to it measurements are performed.
tion. The sensors detect direct and
Inline sensing is used for the evalua-
diffuse reflection for background sig-
tion of direct absorption. The setup
nal treatment. For fluorine tracking
in perpendicular direction considers
absorption measurements in the ul-
the diffuse contributions of absorp-
traviolet and infrared regime are per-
tion, see Fig. 3. The algorithm of the
formed. The received signal is pro-
FT is constantly improved by machine
portional to the content of fluorine
learning (ML) routines. The arrange-
and the thickness of the wax layer.
ment of the multi-sensor setup re-
Thus, thin layers with high fluorine
quires multidimensional optimization
content may result in the same signal
to sharpen its performance [11].
as thick layer with low fluorine con-
Figure 3: Setup of
the sensing device.
Influences on the optical signal In the worst case, the fluorescence
The quality of the optical signal is phenomena are in the range of the
largely determined by the reflectiv- UV absorption. Since the ski base
ity of the surface. This is influenced is usually filled with carbon black or
by the ski base peeling, grinding as nanoparticles, the optical influences
well as waxing, brushing and pol- of these additives also play a role.
ishing. Hand structures have an in- By proper ski preparation an intimate
fluence as well [12]. In addition bond between polymer and wax is
to the directional reflection, there formed [13] which interacts with light
is also a diffuse component caused and sometimes generates a fatty look
by the peeling and grinding ridges of the base. Finally, the influence of
as well as by introduced structures. scattered light should be mentioned,
Furthermore, the quality of the ski which may lead to the general re-
base polymer as well as its color and duction of signal intensities and to
transparency have a significant in- a significant decrease of the signal-
fluence. In some cases, color pig- to-noise-ratio. Thus, a complex pa-
ments fluoresce, which can lead to rameter space is created which can
overlaps with the measured values. only be addressed with the multi-
Open Access Journal – ISSN 2509-9442 – snowstorm-gliding.de 16
sensor approach demonstrated and prepared using the common steps
the self-improving algorithm men- of ski preparation. If the last step
tioned above. of preparation, that means brush-
ing, was performed properly, wax
Information depth and polymer form an indistinguish-
Both the fluorine tracker and FTIR in- able compound with a thickness be-
struments have an information depth tween 1 and 2 μm [15]. Therefore,
of approx. 0.1 to several 10 μm. for calibration, the wax layer thick-
This means that the excitation with ness was considered constant. Im-
infrared light excites absorbing vi- proper preparation results in thicker
brations down to this depth. Thus, wax layers.
an FTIR instrument measures both
the wax layer and the underlying ski Results
base, which can lead to distortions Field tests have so far been carried
in the case of a fluorinated ski base. out in Falun, Östersund and Lenzer-
When excited with ultraviolet radia- heide. For this purpose, 3 devices
tion, the depth of information is sig- were in use. For the tests, ski tech-
nificantly less. With ultraviolet light, nicians from several national teams
valence electrons are excited. Their were asked to prepare skis and make
return to the ground state can occur them available for the FT measure-
without radiation, via fluorescence ments. The tests served to optimize
or via phosphorescence. Due to the the evaluation algorithm. During the
very high absorption coefficients of first tests in Falun, a non-optimized
most polymers in this spectral range, version was used, which still unsat-
the penetration depth is less than isfactorily reflected the condition of
100 nm [14]. Since the FT uses wave- the skis. Based on the information
lengths greater than 250 nm, even provided by the ski technicians, the
shallower penetration depths can be algorithm was refined and a much
expected. Thus, this channel is only improved version was used in Öster-
open to near-surface layers. sund and Lenzerheide. The quality of
fluorine detection is thus dependent
Calibration on the quality of the information pro-
Since optical methods provide lim- vided by the ski technicians. Labora-
ited quantitative information only, tory tests were performed in parallel
the FT must be connected to a stan- to improve the correlation between
dard. The method used is XPS, which preparation condition and FT mea-
in turn provides accurate, energy- surement results.
resolved spectra by using the built-
in gold sample for spectra calibra- Alpine World Cup in Lenzerheide
tion. Wax samples with low fluorine Figure 4 clearly shows that waxes
concentrations and samples of flu- with different fluorine content can be
orinated polymers are used to cali- differentiated. The diagram is based
brate the FT. XPS measurements are on 53 individual measurements. The
performed on these samples, out- error bars underline that the best
putting concentrations as atomic or differentiation is possible for highly-
mass percent. A calibration curve is fluorinated waxes with respect to low-
obtained, which is then stored in the fluorine waxes. The difference be-
FT. Since the physical principles of the tween low-fluorine waxes and waxes
two methods are fundamentally dif- containing no fluorine is not as pro-
ferent, systematic errors are avoided. nounced, thus the error bars nearly
To address the issue that for all op- overlap. However, it has to be men-
tical methods – i.e., the FTIR and FT tioned, that at a certain fluorine con-
– besides concentration of the ob- centration the tribological effect of
served species the signal also de- suppressed capillary action vanishes.
pends on the thickness of the mea- This threshold is located in the lower
sured layer, calibration samples were border of the low-fluorine box.
Open Access Journal – ISSN 2509-9442 – snowstorm-gliding.de 17
Figure 4: FT signals
of 53 pairs of alpine
skis during the Lenz-
erheide worldcup in
March 2021.
Summary the competition, but an indication per
The FT is an instrument based on the red/green signal is used for decision
latest physical measurement tech- making.
nology and data processing. The
device is self-learning, i.e., the de- Acknowledgment
tection of fluorine improves with in- The author would like to thank per-
creasing number of measurements. sons involved in the field tests.
The handling is easy and the light- In alphabetical order the acknowl-
exposure time per ski is in the range edgment goes to Max Cobb (IBU),
of seconds. By connecting the FT to Lars Karlöf and Atle Skårdal (FIS).
the XPS method, absolute measure- My acknowledgment further goes to
ments are possible with atomic con- Prof.(em.) Dr. Jürgen A. Schäfer for
centrations. However, these results his advice in the area of surface sci-
are not published during the tests at ence.
About the Author
Matthias Scherge is a professor of tribology. This is the
science of friction, wear and lubrication. Prof. Scherge
heads the Fraunhofer MikroTribologie Centrum, teaches at
the Karlsruhe Institute of Technology and manages Team
Snowstorm. He also advises the Nordic Paraski Team Ger-
many and several national and international athletes on
scientific and technical issues.
References
[1] Christos Stamboulides, Peter Englezos, and Savvas G Hatzikiriakos. Ice
friction of ultra-high molecular weight polyethylene: The effects of
fluorine additives and plasma (PECVD) treatment. Tribology
International, 57(0):177–183, jan 2013.
Open Access Journal – ISSN 2509-9442 – snowstorm-gliding.de 18[2] Roman Böttcher, Marc Seidelmann, and Matthias Scherge. Sliding of
uhmwpe on ice: Experiment vs. modeling. Cold Regions Science and
Technology, 141:171 – 180, 2017.
[3] Martin Schlabach. POPFREE Ski Goes Global. personal communication,
2020.
[4] Baard Ingegerdsson Freberg, Raymond Olsen, Syvert Thorud, Dag G.
Ellingsen, Hanne Line Daae, Merete Hersson, and Paal Molander.
Chemical Exposure among Professional Ski Waxers— Characterization
of Individual Work Operations. The Annals of Occupational Hygiene,
57(3):286–295, 11 2012.
[5] Helena Nilsson, Anna Kaerrman, Anna Rotander, Bert van Bavel, Gunilla
Lindstroem, and Hakan Westberg. Professional ski waxers exposure to
PFAS and aerosol concentrations in gas phase and different particle
size fractions. Environ Sci Process Impacts, 4:814–822, 2013.
[6] INTERNATIONAL SKI FEDERATION. Summary of the FIS Council Meeting,
23rd November 2019, Konstanz / Constance (GER). FIS Bulletin, 2019.
[7] S. Rohm, S.H. Unterberger, M. Hasler, M. Gufler, J. van Putten,
R. Lackner, and W. Nachbauer. Wear of ski waxes: Effect of
temperature, molecule chain length and position on the ski base.
Wear, 384-385:43–49, 2017.
[8] Martin Schlummer, Ludwig Gruber, Dominik Fiedler, Markus Kizlauskas,
and Josef Müller. Detection of fluorotelomer alcohols in indoor
environments and their relevance for human exposure. Environment
International, 57-58:42 – 49, 2013.
[9] Isabelle Rogowski, Jean-Yves Gauvrit, Didier Léonard, and Pierre
Lanteri. Typology of the gliding waxes in cross-country skiing:
Comparison between classifications based on the chemical composition
and those based on the physical and physicochemical properties. Cold
Regions Science and Technology, 43(3):140–149, dec 2005.
[10] Isabelle Rogowski, Didier Leonard, Jean-Yves Gauvrit, and Pierre
Lanteri. Influence of fluorine-based additive content on the physical
and physicochemical properties of ski gliding wax. Cold Regions
Science and Technology, 49(2):145–150, aug 2007.
[11] Leon Bottou, Frank E. Curtis, and Jorge Noceda. Optimization Methods
for Large-Scale Machine Learning. SIAM Review, 60:223–311, 2018.
[12] Matthias Scherge and Christian Winker. Zur Wirkung von
Strukturgeräten. Gliding, 1:1–5, 2020.
[13] Matthias Scherge. Wachs oder kein Wachs – Das ist hier die Frage.
Gliding, 1:1–3, 2016.
[14] C. W. Emmons and Alexander Hollaender. The action of ultraviolet
radiation on dermatophytes. ii. mutations induced in cultures of
dermatophytes by exposure of spores to monochromatic ultraviolet
radiation. American Journal of Botany, 26(7):467–475, 1939.
[15] Matthias Scherge. Wax or no wax – That is the question. Gliding, 1:1–3,
2016.
Open Access Journal – ISSN 2509-9442 – snowstorm-gliding.de 19You can also read
