Synthesis and application of additives based on trifluoroethoxy-cyclo-phosphazene into polymer nanofibers

Article abstract of DOI:10.1016/j.tet.2020.130999

This paper is focused on the synthesis and application of phosphazene derivatives, hexakis(2,2,2–trifluoroethoxy)-cyclo-triphosphazene and tetrakis(2,2,2-trifluoroethoxy)-diamino-cyclo-triphosphazene. These compounds were used as additives into some common polymers. As the synthesized compounds themselves show enhanced hydrophobicity, the nanofibers, made from them by electrospinning technology, should exhibit the enhanced hydrophobicity, too. It was found that hydrophobic properties of fibers are influenced by number of trifluoroethoxy groups bound to the phosphazene cycle. The use of these modified nanofibers especially in surface protection is expected. The result of the use can be the creation of efficient, removable, compact protect layer for long-term objects storing.

Full text of DOI:10.1016/j.tet.2020.130999

Journal Pre-proof
Synthesis and application of additives based on trifluoroethoxy-cyclo-phosphazene
into polymer nanofibers
Radka Kopecka, Milan Alberti, Jiri Prihoda, Zbynek Vorac, Lucie Zarybnicka
PII:
S0040-4020(20)30095-8
https://doi.org/10.1016/j.tet.2020.130999
TET 130999
DOI:
Reference:
To appear in: Tetrahedron
Received Date: 13 October 2019
Revised Date: 29 January 2020
Accepted Date: 31 January 2020
Please cite this article as: Kopecka R, Alberti M, Prihoda J, Vorac Z, Zarybnicka L, Synthesis and
application of additives based on trifluoroethoxy-cyclo-phosphazene into polymer nanofibers,
Tetrahedron (2020), doi: https://doi.org/10.1016/j.tet.2020.130999.
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. Please note that,
during the production process, errors may be discovered which could affect the content, and all legal
disclaimers that apply to the journal pertain.
© 2020 Published by Elsevier Ltd.

Synthesis and Application of Additives based on Trifluoroethoxy-cyclo-phosphazene into Polymer
Nanofibers
Radka Kopeckaa*, Milan Albertib, Jiri Prihodaa, Zbynek Voracb, Lucie Zarybnickac
aMasaryk University, Department of Chemistry, Faculty of Science, Kamenice 753/5, 611 37 Brno, Czech
Republic
bMasaryk University, Department of Physical Electronics, Faculty of Science, Kotlarska 267/2, 611 37 Brno,
Czech Republic
cCollege of Polytechnics Jihlava, Department of Technical studies, Tolsteho 16, 586 01 Jihlava, Czech Republic
Correspondence to: Radka Kopecka; e-mail address: radkabacovska@centrum.cz, tel. +420 728379398
Abstract
This paper is focused on the synthesis and application of phosphazene derivatives, hexakis(2,2,2–
trifluoroethoxy)-cyclo-triphosphazene and tetrakis(2,2,2-trifluoroethoxy)-diamino-cyclo-triphosphazene. These
compounds were used as additives into some common polymers. As the synthesized compounds themselves
show enhanced hydrophobicity, the nanofibers, made from them by electrospinning technology, should exhibit
the enhanced hydrophobicity, too. It was found that hydrophobic properties of fibers are influenced by number
of trifluoroethoxy groups bound to the phosphazene cycle. The use of these modified nanofibers especially in
surface protection is expected. The result of the use can be the creation of efficient, removable, compact protect
layer for long-term objects storing.
Keywords: hydrophobicity, modification, nanolayers, polymer additives, polyphosphazenes, media filter
1. Introduction
Recently, various nanotechnologies have been used in many research fields and applications.
Nanotechnology utilizes organic or inorganic materials having either monomeric or polymeric character and
taking the form of nanoparticles or nanofibers. The numerous “nano” materials can be further chemically
modified, i.e. their molecules can be functionalized with a variety of side groups. This modification leads to
newly created bonds in the basic chemical structure, which can significantly modify the properties of the final
product [1].
1

Another way to achieve nanofiber modification (i.e. without creating new bonds) simply consists in mixing
single components and their further processing. In this article, the behaviour of organic polymers mixed with
phosphazene additives is described. Polymeric nanofibers were made by electrospinning technology and their
properties were determined. Usually, the resulting basic polymer is expected to preserve its original structure
where certain physical properties (such as hydrophobicity, thermal stability) can be slightly changed due to the
presence of an additive [2].
In our research, we used different types of commercially available “matrix” polymers and above mentioned
phosphazene derivatives that contain non-polar organic side groups and that were applied as additives into basic
polymers. We have chosen the following polymers PA6, PS, ASA, SAN and PESU, which find application as
materials for filtration [3-11]. Deep bed filtration is an economic, efficient oily wastewater treatment technology
that is particularly suited to removing oil pollutants with diameters of less than 10 ꢀm [12].
The cyclic or linear phosphazene molecules, which this article deals with, are composed of P-N backbone
containing two types of substituents - trifluoroethoxy and amino groups, bound directly to the phosphorus atoms.
It has been known for long time that many compounds containing phosphorus and nitrogen atoms can be
effective as fire suppressants [13]. This behaviour is usually more pronounced in the case of the phosphazene
backbone. Moreover, the bonding of halogen atoms (mainly fluorine ones) to the phosphazene compounds
presumably increases their flame extinguishing capability. It is also well known that trifluoroethoxy groups
could significantly increase the hydrophobic character of the whole cyclo-phosphazene molecule [14] as well as
the hydrophobic character of the polymer further modified by the phosphazene additives [14-15].
Currently, nanofibers can be produced on an industrial scale from a plenty of commercial polymers in
number of ways, including electrospinning technology.
The electrospinning technology utilizes high electrostatic field to fiberize substances from their solutions or
melts. For this purpose, the device contains a positively charged cylindrical electrode (placed into the basin with
a fiberized material) and a negatively charged grounded electrode, called the collector. If the electrostatic forces
are greater than the surface tension of a fiberized polymer solution, the charged liquid is ejected from the
cylindrical electrode towards the collector. The solvent evaporates during fiber formation and the resulting fibers
split and elongate themselves. The fibers are deposited on the supporting material (a layer of textile or another
polymer) that the collector is covered with. The scheme of the device used is shown in Figure 1 [15-17].
2

Fig.1: Scheme of the electrospinning apparatus -NanospiderTM
.
Nanofibers have great potential in a wide range of applications mainly due to their very large surface area.
Our effort to modify the surface hydrophobicity of some polymer fibers by using phosphazene additives (which
contain trifluoroethoxy groups) could lead to a wider utilization of these materials in new technological
applications, e.g. the modified nanofiber layers could be used for media filter [18]. The wettability of the filter
media is a key factor affecting the oil removal efficiency of this technology. Generally, the oil removal
efficiency increases as the lipophilicity and hydrophobicity of the filter media increase [19]. Filter media of such
modified polymer nanofibres with hydrophobic properties could exhibit the ability to not adhere to undesirable
pollutants. The filter medium would have this effect due to the improved filtration capability and longer service
life.
3

2. Experimental part
2.1. Synthesis of hexakis(trifluoroethoxy)-cyclo-triphosphazene (3) and tetrakis(trifluoroethoxy)-diamido-
cyclo-triphosphazene (4)
The common mechanism of polyphosphazene synthesis is the ring-opening polymerization of hexachloro-
cyclo-triphosphazene (1) at 250 °C to form poly(dichlorophosphazene) (5).
The functionalization of polymer (5) as well as trimer (1) is commonly followed by the nucleophilic
substitution reaction, e.g. by using sodium trifluoroethoxide in THF [20]. However,
poly[bis(trifluoroethoxy)phosphazene] (synthesized from 5) can be hardly joined with other materials due to its
high surface hydrophobicity. Thus, its practical use is not usually so simple. Therefore, we synthesized
compounds
- hexakis(2,2,2-trifluoroethoxy)-cyclo-triphosphazene (3) and tetrakis(2,2,2-trifluoroethoxy)-
diamido-cyclo-triphosphazene (4) - which are possible to combine with other commercially available polymers.
Thus, these compounds (3 and 4) can improve especially hydrophobic properties of the final product.
The easily synthesized hexachloro-cyclo-triphosphazene, P3N3Cl6 (1), is commonly used as the starting
material for many other substitutions [21-27]. The course of the phosphazene derivatives nucleophilic reaction
with a nucleophile reagent strongly depends on the reaction conditions, such as the molar ratio of reactants, the
solvent and temperature.
Alcoholysis is one of the common nucleophilic substitution reaction types. The synthesis of fluoroethoxy-
cyclo-triphosphazene derivatives was carried out according to Allcock and Schmutz [28]. The course of the
reaction is shown in Fig. 2. Compound 2 was prepared by ammonolysis using the parent compound 1 and gas
ammonia in diethyl ether [29]. Compounds 3 and 4 were first reported in late 20th century (Fig. 2) [30-32].
4

Fig.2: Fluorophosphazene derivative syntheses scheme.
All syntheses and sample preparations were performed using the Schlenk technique in a dry nitrogen
atmosphere. Solvents, such as diethyl ether (Sigma Aldrich, p.a.) were purified and dried according to
procedures described by Perrin and Armarego [33].
For preparation of sodium trifluoroethanolate, small pieces of metallic sodium were washed first by dry
diethyl ether and then slowly added to the cooled solution of 2,2,2-trifluoroethanole in dry diethyl ether, placed
in a Schlenk tube, equipped with a screw cap. Then, the solution of either 1 or 2 in dry diethyl ether was slowly
added dropwise to the stirred and cooled solution of sodium trifluoroethanolate and the mixture was stirred for 8
(for compound 1) or 24 (for compound 2) hours at room temperature. The starting composition of the reaction
mixtures can be seen in Tab.1.
5

Table 1: Amounts of starting compounds and solvents used in the synthesis of products 3 and 4.
Product
Na
CF3CH2OH
10.33 g
1
2
(C2H5)2O
Hexakis(trifluoro-ethoxy)-cyclo-
triphosphazene (3)
2.38 g
5.96 g
-
80 ml
(0,11 mol)
7.415 ml
(0,10 mol)
1.04 g
(0,02 mol)
Tetrakis(trifluoro-ethoxy)-
0.24 g
-
0.58 g
60 ml
diamido-cyclo-triphosphazene (4) (0,01 mol)
0.75 ml
(1,88.10-3 mol)
(0,01 mol)
After cooling the reaction mixture, the by-product (sodium chloride) was removed by filtration over the
Schlenk frit (S4) filled with 3 sheets of paper filter. Then, the filtrate was purified by heating with active
charcoal in dry diethyl ether. Partially substituted derivatives were removed in this process at the same time. The
solvent was then partly evaporated and the remaining liquid, containing products 3 or 4, was filtered using a
Schlenk frit S4. Product 3 was a viscous liquid from which transparent crystals were gained at a low temperature
(around -5 °C), the melting point was 47.3-48.8 °C. Product 4 was also a viscous liquid and after removing
excess solvent by using vacuum distillation, whitish crystals with the melting point 49.9-51.6 °C were gained.
The percentage yield was 87% for product 3, 83% for product 4.
Both crystalline products (3, 4) were filtered off. After their identification (see below), the compounds were
used as additives to commercial polymers for creating nanofiber layers.
2.2. Preparation of nanofiber layers
Three types of nanofibers, which differ in their composition, were formed.
1) The commercial polymers - PS, PA6, PESU or copolymers - ASA and SAN - were fiberized on their
own (Fibers A).
2) The aforementioned polymers with additive 3 (Fibers B).
3) The same commercial polymers with additive 4 (Fibers C).
Not only the influence of the additives presence alone, but the additive amount in the spinning mixture was
also investigated. Therefore, experiments were carried out with 10, 25 and 50 wt.% of both additives
(compounds 3 or 4).
6

The composition of fiberized mixtures
The commercial polymers/copolymers (PA6, PS, ASA, SAN, PESU), produced by BASF, were dissolved in
one of the following diluents - trifluoroacetic acid (TFA), a mixture of acetic and formic acids (AFA) or
dimethylformamide (DMF), (Sigma Aldrich, Czech Republic, purity min. 99%). The solvents were purified by
distillation before use.
Preliminary experiments showed that low concentrations of additives have negligible effect on the properties
of polymer fibers [34]. Therefore, the content of additives was examined at 10, 25, and 50 wt.%. It should be
stated that the unfiberized residual was found as a white powder on the rod electrode after performing the
spinning process in the case of 50 wt.% of 3 or 4. The 31P NMR spectroscopy proved the presence of 3 or 4
additive in this powder. Therefore, only the results provided by nanofibres containing either 10 or 25 wt.% of
additives are presented in this article. No residual powder was observed on the electrode after the spinning
process and thus it can be assumed that the whole amount of additive was fiberized together with a commercial
polymer. The final composition of the total fiberized mixture (i.e. commercial polymer + 3 or 4 additive +
solvent) was approximately kept the same, so fiberized mixture containing PA6 in TFA or AFA contained
approximately 12 wt.%, the mixtures containing PS, ASA or SAN in DMF 25 wt.% and the mixture of PESU in
DMF 20 wt.% of solid parts. These compositions were chosen on the basis of our previous experiments [34].
For following fiberizing experiments, the solution of dissolved polymers was divided into three parts. The
first portion of the sample was fiberized on its own, i.e. nanofibers were created only from a single
polymer/copolymer (Fibers A). The second portion was mixed with 3 as an additive (Fibers 10B, 25B, 50B)
either in 10, 25 or 50 wt.%. The compound 4 was added in 10, 25 and 50 wt.% as an additive into the third
portion of the sample, (Fibers 10C, 25C, 50C).
All three parts were homogenized for 30 minutes using an electromagnetic stirrer before electrospinning.
Fiberizing of polymeric mixtures (Fibers A, B, C)
Nanofibers were formed from the cylindrical electrode (see Fig. 1) using the Elmarco Nanospider™ NSLAB
500 pilot plant unit. Polypropylene foil (PP foil) was used as the supporting material covering the grounded
electrode. In the electrospinning process, 20 ml of fiberized solution was placed into the basin equipped with a
constantly rotating cylindrical electrode (see Fig. 1).
7

The distance between the electrodes was 150 mm, the voltage was 64.5 kV, the current was around 0,001 mA
and the rotation speed of electrode was 10.2 rpm. The above mentioned parameters were kept at the same level
for all samples.
The experimental conditions were as follows - temperature 25.5 °C, relative humidity 27 %, the shift speed
of the supporting PP foil was 1 mm/min. Each sample was fiberized at the length of 30 cm, the deposition
process was repeated at least 10 times, there and back in the length (30 cm), to create an approximately 50-
micrometer-thick nanolayer.
Then the nanofibers were removed from Nanospider together with the supporting PP foil and were allowed to
dry at 25 °C overnight in order to evaporate the rest of the solvent. The samples were placed in plastic bags for
further analysis.
2.3. Methods for analysis
The chemical composition of synthesized phosphazene additives (3, 4) as well as of nanofibers was
determined by physico-chemical methods. The synthesized phosphazene additives (3, 4) were characterized by
31P NMR spectroscopy, MS spectrometry and X-ray diffraction. The presence of these additives in the fiber
layers was confirmed by ATR-FTIR and by 31P NMR spectroscopies. The macroscopic structure of the fibers
was investigated using the SEM and their hydrophobicity was determined using the SEE system analysis. The
thermal stability of nanofibers was investigated by TGA and DSC analysis.
2.3.1. 31P Nuclear magnetic resonance (NMR) spectroscopy
31P NMR spectra were measured from solutions of the reaction mixture or dissolved nanotextile. A BRUKER
AVANCE DRX 300 instrument at the frequency of 31P 202.46 MHz was used for recorded 31P NMR spectra (δ,
ppm). 85% solution of H3PO4 was used as the external standard. The spectra were measured in the coaxial NMR
cuvette system. The samples were sealed in Simax tubes (diameter 4 mm), inserted in NMR cuvettes (diameter 5
mm) filled with D2O (external lock). The MestRe Nova NMR program was used to evaluate the spectra.
2.3.2. Fourier transform infrared (ATR-FTIR) analysis
The Bruker Vertex 80V spectrometer was used for measuring the attenuated total reflectance of ATR-FTIR
spectra. The solid samples were measured in attenuation total reflection mode using a diamond crystal with the
spectral range calibration from 4000 to 600 cm-1. The sample compartment was evacuated to 2.51 hPa. Each
spectrum was scanned a hundred times. Bruker OPUS software was used for spectrum evaluation and the
8

baseline correction was done by “rubber-band”. The characteristic bond vibration of the phosphazene cycles
occurred mainly at 1210.32 cm-1.
2.3.3. Mass spectrometry (MS)
MS was performed using an Agilent Technologies - GC Model 6890N with MSD Model 5975B. The heating
zone temperature ranged 50 - 250 °C, the column heating rate was 40 °C/min up to 250 °C. The injector
temperature was maintained for 5 min at around 250 °C. Detector temperature was 200 °C. The column pressure
was 150 kPa. The value of the excitation energy used for electron impact ionization was 70 eV.
2.3.4. X-ray diffraction
The single-crystal X-ray diffraction experiments were performed using the Rigaku diffraction system
(MicroMax007HF DW rotating anode source, multilayer optic, partial χ axis goniometer and Saturn 724+ HG
detector) with molybdenum radiation (Kα, λ = 0.71075 Å). Crystal screening, strategy calculation and data
collection were performed using Crystal Clear software [35]. The data reduction was performed using
CrysAlisPro software [36]. Both structures were evaluated and the results were refined (full matrix least-squares
refinement on F2) using SHELXT and SHELXL programs from SHELX system [37].
2.3.5. Scanning electron microscope (SEM) analysis
The SEM images of the nanofibers were made using the Tescan MIRA3 scanning electron microscope. The
nanolayer was fitted onto the pad using a sticky tape and its surface was covered by a 30 nm gold layer. The
samples were measured using the secondary emission mode, the depth regime of 15 kV, the working distance
(WD) of 10 mm and 20,000 x magnification. Contrast-enhancing colouring was used for contrast visualization.
The fiber size was determined using ImageJ software with manually located fiber borders [38-39].
2.3.6. Surface Energy Evaluation System analysis (SEE system analysis)
The contact angle between free surface of nanolayers and water drop characterizes the extent of
hydrophobicity [40]. The contact angle measurement of nanolayers was based on the standard “EN 828 -
Adhesives - Wettability - Determination by measurement of contact angle and surface free energy of solid
surface” [41]. Approximately 2 cm2 of the central part of nonwoven layer was placed on the microscope pad and
constant volumes (1ꢀl) of distilled water were manually dropped on the nanolayer surface. The snap of the water
9

drop was made, the contact places between the drop and the supporting textile were manually marked and the
contact angle was determined. The value of contact angle was determined by tangent method using Software 7.0.
The only supporting pure PP foil was also tested for its influence on the value of contact angle. No significant
differences between surface contact angles measured in the pure nonwoven textiles and nonwoven textiles fixed
on the supporting PP foil were determined. Therefore, the measurements were carried out with the samples
deposited on the PP foil, for easier manipulation.
2.3.7. Thermogravimetric analysis (TGA)
TGA was carried out using a TA Instruments TGA Q500 device. The samples of pure synthesized additives
of 29.879 mg (additive 3) and 11.699 mg (additive 4) were investigated at a heating rate of 10.0 °C/min, in the
temperature range 20-1000 °C. The air was used as a purge gas at a flow rate of 60.0 ml/min. The minimum
thermal decomposition temperature (Tda) was determined.
2.3.8. Differential scanning calorimetry (DSC)
Differential scanning calorimetry (DSC) measurements were performed on NETZSCH STA-449C Jupiter
with SiC type furnace. The scales were kept at a constant temperature of 30 °C and protected from heat and
furnace exhaust with a stream of nitrogen. Samples were measured in corundum crucibles, measured under
nitrogen atmosphere (purity 5.0). The gas flow rate through the furnace was 100 cc/min, the heating rate was 10
K/min, in the temperature range 25-500°C. The resulting thermograms were processed and evaluated using the
Netzsch Proteus software.
3. Results and discussion
3.1. Characterization of nanofibers and their phosphazene precursors
3.1.1. 31P NMR analysis
31P NMR analysis confirmed the purity of synthesized phosphazene precursors 3 and 4. Compound 3 is
characterized by multiplet AB2 with the chemical shift around δ(M) 17.46 ppm. Compound 4 is described by
singlet with the chemical shift δ(S) 17.18 ppm in A3 spectra type. The 31P NMR spectra of OCH2CF3 fully
substituted product 4 is accompanied by signals of some other tetrasubstituted derivative, which was determined
as P3N3(OCH2CF3)4Cl2, and characterized by spectra AB2 with doublet (-PCl2) and triplet (-P(OCH2CF3)2) with
the chemical shifts δ(D) = 15.12 ppm and δ(T) =27.32 ppm.
10

The 31P NMR spectra of dissolved nanotextile were measured in the same solvent that was used for
electrospinning. For confirmation, the characteristic signals for PN rings (for additive 3, δ(M) 17.46 ppm, and
for additive 4, δ(S) 17.18 ppm) were searched for in the spectra. Even though the quality of the spectra was poor,
because of a low concentration of 3 or 4 in dissolved nanotextiles, minor signals belonging to phosphazene
additives 3 or 4 were observed in the spectra. The quality of spectra was then a little bit improved using a higher
scan number (500 scans).
3.1.2. Nonwoven textile ATR-FTIR analysis
ATR-FTIR analysis was performed for both pure additives 3 and 4 (Fig. 3) as well as spinned nonwoven
textile. The vibrations at around 1200 cm-1, belonging to the phosphazene ring, were observed for both 3 and 4
samples, whereas the vibrations at around 3250 to 3450 cm-1, belonging to -NH2 groups bound to the
phosphazene ring [42], were detected just in the case of presence compound 3.
Fig. 3: ATR-FTIR spectra
11

. In the ATR-FTIR spectra of nanofibers, used for determining the presence of additives 3 or 4 in these fibers,
containing additives 3 (green lines) or 4 (red lines) combined with polymers - PS, ASA and SAN (Fig. 5B, C,
D), vibrations at around 1200 cm-1, belonging to the phosphazene cycle, were observed. No vibrations at about
the same wavenumbers were found in the pure polymer (blue lines). In the case of PA6 and PESU, it was
difficult to determine these vibrations (Fig. 5A, E), because the vibrations at around 1200 cm-1, belonging to
phosphazene additives 3 or 4, overlapped with those of the pure polymers.
3.1.3. Mass spectrometry analysis of phosphazene additives
The data from direct inlet mass spectroscopy was used for the characterization of 3, 4. Some identified
fragments for compounds 3 and 4 are listed in the Tab. 2. Thus, the presence of additives 3 and 4 was uniquely
confirmed.
Table 2: MS fragments of 3, 4.
m/z
m/z
fragments
theoretical experimental
[P3N3Cl6]+
347.66
531.07
548.08
566.52
630.10
646.10
665.55
691.12
710.13
729.13
308.79
435.95
531.07
544.02
348
532
547
567
629
647
665
691
709
729
308
435
532
544
[P3N3(OCH2CF3)4]+
[P3N3(OCH2CF3)4OH] +
[P3N3(OCH2CF3)4Cl] +
[P3N3(OCH2CF3)5] +
3
[P3N3(OCH2CF3)5O] +
[P3N3(OCH2CF3)5Cl] +
[P3N3(OCH2CF3)5 (OCH2CF)] +
[P3N3(OCH2CF3)5 (OCH2CF2)] +
[P3N3(OCH2CF3)6] +
[P3N3Cl4(NH2)2]+
[P3N3Cl2(OCH2CF3)2 (NH2)2]+
[P3N3(OCH2CF3)4]+
4
[P3N3(OCH2CF3)4(NH2)OH] +
12

[P3N3(OCH2CF3)3(NH2)2] +
[P3N3(OCH2CF3)2(NH2)2OH] +
[P3N3(NH2)2(OCH2CF3)4] +
464.18
382.56
308.80
464
382
309
3.1.4. X-ray diffraction
The crystal structures of 3 (Fig. 4) and 4 (Fig. 5) were confirmed by X-ray diffraction. Both compounds are
monoclinic, space group P21, Z=16 for compound 3 and space group P21/C, Z=8 for compound 4. The hexa-
membered P3N3 rings are slightly distored which is comparable with other hexa-membered phosphazene rings
[35]. There are eight individual molecules of substances 4 arranged into layers in monoclinic lattice always
mutually facing with -CF3 groups, while this phenomenon was not observed in compound 3 where only 2
individual molecules are in the lattice.
Fig. 4: The crystal structure of 3.
Fig. 5: The crystal structure of 4.
13

Crystals of sample 3 were of poor quality (substantially anhedral crystals with non-consistent behaviour
under polarized light). However, it was possible to find an individual pattern with reasonable diffraction. The
poor quality of crystal structure 3 may be the result of the presence of eight symmetrically independent
molecules in a lattice cell of 3, and, thus, a relatively large quantity of parameters was difficult to specify. Some
intermolecular weak interactions and large ADPs caused by vibrations of CF3 groups also played a role in the
process. The selected crystal 3 did not diffract up to atomic resolution. Reflections with dHKL< 0.90 Å were
omitted from the refinement. Rigid bond restraints were applied to all atoms. Geometrical similarity restraints
were applied to all trifluoroethoxy moieties. ADP similarity restraints were applied to all atoms closer than 1.0 Å
(affecting only 3 disordered trifluoroethoxy fragments). Hydrogen atoms were placed at calculated positions and
refined as riding, with their Uiso set to 1.2 Ueq of the carrier atom.
The refinement of structure 4 was, on the other hand, relatively straightforward. Three out of eight
(symmetrically independent) trifluoroethoxy fragments were refined as disordered (effectively splitting only
trifluoromethyl moiety and hydrogen atoms of methylene groups); these were treated by similar restraints, as
used in refinement of 3. Hydrogen atoms in the methylene groups were placed at calculated positions and refined
as riding, with their Uiso set to 1.2 Ueq of the carrier atom, other hydrogen atoms (bound to nitrogen atoms) were
geometrically unrestrained, only the N–H bond lengths were restrained to be equal to 1.01 Å and Uiso values of
these hydrogens were set to 1.2 Ueq of the nitrogen carrier atoms. Some selected crystallographic parameters of
structures 3 and 4 are listed in Tab. 3.
Table 3: Selected crystallographic parameters of structures 3 and 4.
Compound
3
4
Molecular formula
C12H12F18N3O6P3
729.16
C8H12F12N5O4P3
563.14
Mr
Space group
P21
P21/c
Calculated density
1.908 Mg m−3
21.9883(2) Å
19.9631(2) Å
23.2493(3) Å
95.866(1)°
1.879 Mg m−3
9.1247(1) Å
25.9847(2) Å
16.8915(1) Å
96.160(1)°
a
b
c
β
14

V
10151.95(19) Å3
16
3981.89(6) Å3
8
Z
µ
0.40 mm−1
120 K
0.44 mm−1
120 K
T
Measured reflections
Independent reflections
Rint
77914
42945
29071
7519
0.046
0.019
θ range (refinement)
R[F2> 2σ(F2)]
wR(F2)
2.7°– 23.3°
0.147
3.2 °– 25.7°
0.029
0.435
0.076
S
1.83
1.05
Parameters / restraints
ꢁρmax
3191 / 18232
2.52 e Å−3
−1.26 e Å−3
712 / 960
0.42 e Å−3
−0.38 e Å−3
ꢁρmin
3.2.3. The surface analysis of nanofibers
In Fig. 6, the SEM images of the nanofiber layers are shown. It follows from the SEM snaps that the
nanofibers made from the pure PA6 and SAN polymers, as well as the nanofibers created from PA6 and SAN
together with either additive 3 or additive 4 (in the amount of 10 wt% of the additive), have relatively
homogenous smooth surface without beads or other irregularities.
No fiber orientation was observed there. Thus, these homogenous nonwoven textiles could be used for filter
materials. On the contrary, the fibers made from PESU and especially ASA do not create homogenous layers.
Actually, there was a predominance of beads over fibers by the sample consisting of ASA (see Fig. 6). This
observation follows from the snaps taken from samples containing both polymers (ASA and PESU).
15

Fig. 6: Results of fibers SEM analyses.
It is known that electrospinning process can be accompanied also by electrospraying (i.e. the creation of
drops, which transform into beads rather than fibers, after the solution evaporates) [43-45]. Therefore, the
formation of nonhomogeneities can be explained by simultaneous course of electrospinning and electrospraying.
In the case of the layer spinned from ASA, individual irregular beads are usually created, whereas in the case of
PESU, the beads appear only as defects on fibers. Therefore, it can be assumed that a nanolayer containing ASA
(using aforementioned concentrations) was more likely created by the electrospraying process rather than by
electrospinning [46]. This leads to a lower homogeneity of this layer.
Although, there are no beads in the nanofibers made from PS, the fiber surface was scarred (not smooth) and
thus the nanolayer doesn´t seem to be homogenous. Scarring of the surface can be explained as follows: the
fiberized mixture is rapidly charged and many Taylor cones (= the place where the nanofiber is ejected from the
polymer drop on the electrode) are formed. It causes the individual fibers to stick together before their deposition
16

on the supportive PP foil. This grouping of individual fibers then moves together towards the collector. The
solvent is also difficult to evaporate due to a thicker layer or clusters of stacked fibers. If the solvent evaporates
from the fibers, the cracking of fiber surface was observed
The same results were obtained using 25 wt.% of additives 3 or 4 because the final concentration of the
whole fiberized mixture was kept by the same level for all fiberizing materials (i.e. by the samples containing 10
wt.% as well as 25 wt.%). The fibers created from the mixture with 50 wt.% of additives were not more
examined because of the uncomplete fiberizing process. Only a slightly more irregularities were observed by the
fibers containing 25 wt.% of additives in a comparison with fibers containing 10 wt.% of additives, these fibers
can be still considered as homogenous layers.
The influence of the presence of additive 3 or 4 on the average fiber width was also investigated. Ten fibers
were chosen from the nonwoven textile and the borders of the fibers were manually marked on the snap. The
mean value was calculated from individual widths (Tab. 4). The increase (+)/decrease (-) of the fiber width
(mentioned as how many times bigger/smaller the fiber width is) was compared in fibers containing only
commercial polymer (Fibers A) and in fibers containing a polymer and either 3 or 4 (Fibers B and C). The higher
is amount of additive, the bigger fiber width was observed. Thus, only width values, gained from fibres
containing 3 or 4 in 10 wt.% concentration, are compared and listed in Tab. 4. The results for fibre width for
fibres containing 25 wt.% of additives were increased only by approximately 3 %.
Table 4: The comparison of fiber widths determined by SEM analysis.
MW (µm)
Polymer
Polymer + additive
Pure polymer
(A)
* (A-10B)
-0.14
2.85
*(A-10C)
-0.17
3.11
10B
0.39
1.34
0.10
---
10C
0.42
1.08
0.33
0.13
0.56
PA6
PS
0.25
4.19
PESU
ASA
SAN
0.06
-0.04
---
-0.27
---
---
0.45
0.8
-0.35
-0.11
* increase (+)/decrease (-) of the fiber width
17

It follows from the values of ASA that the differences between the fiber width of Fibers A and Fibers either
10B or 10C were not possible to determine for ASA, because the fiberizing of A was not successful. The slightly
increased values of fibres width 10B and 10C on the contrary to A can be caused by higher electrical
conductivity of spinning mixture containing 3 or 4 additives. The only measurable difference (increase) which
should be mentioned was observed for the samples created from PS (and it is caused by stacked fibers).
For an overview, see Fig. 7, where fibre width size distribution for all fibre samples are listed. In all cases,
they have the lowest values of fibre width for pure polymeric materials. For PS and ASA materials it was not
possible to evaluate values of fibre width for unhomogeneity.
Fig. 7: Fiber width size distribution for all fibre samples.
The behaviour of water drops, being in contact with a nanolayer surface, was determined by contact angle
measurement using the SEE-system. Tab. 5 lists the values of the contact angles for all free fiber types (Fibers A,
10B, 25B, 10C, 25C). The effect of additives was determined: the increase (+)/decrease (-) of the contact angle
(mentioned as how many times bigger/smaller the contact angle is). All samples were measured 30 times with
the maximum value difference 0.9 and their arithmetic mean was determined and reported in Tab. 5.
18

Table 5: The comparison of contact angle values.
water
contact
angle (°)
PA6
pure
polymer
(A)
polymer + additive
*(A-10B)
*(A-25B)
*(A-10C) *(A-25C)
10B
25B
10C
25C
29.9
32.3
38.3
-2.4
-3.3
-14.3
-0.4
-6.2
-8,4
-6,6
60.5
64.8
-30.6
-5.3
-34,9
-12,4
-27,4
-41,6
-56,8
PS
114.9
101.3
80.2
118.2 121.5
115.6 118.9
120.2 127.3
121.1 128.7
119.8 121.8
142.2 146.5
PESU
ASA
-17,6
-11,1
-8,8
-19.8
-39.6
-52.5
80.6
95.9
91.3
98.5
SAN
89.7
* increase (+)/decrease (-) of the contact angle
The data in Fig. 8 show how the values of water contact angle increase in the presence of additives
containing a greater number of –OCH2CF3 groups.
A lower wettability of nanofibers where the more hydrophobic substituents (CF3CH2O- groups) are bound to
the phosphazene backbone was determined. It must also be stated that the hydrophobic or hydrophilic character
of the commercial polymers themselves plays a significant role. According to the size of the contact angle, we
can divide the commercial polymers used into two basic groups - hydrophilic polymers (such as PA6, ASA,
SAN) and hydrophobic ones (PS, PESU). The results shown in Fig. 7 and 8 that originally hydrophilic polymers
- PA6 (29.9°), ASA (80.2°) and SAN (89.7°) - exhibit a more rapid increase in their water repellency than the
originally hydrophobic polymers PS (114.9°) and PESU (101.3°). What is more, some of the originally
hydrophilic polymers such as ASA and SAN change into hydrophobic substances after applying an additive
(substituents with -OCH2CF3 groups). This hydrophobicity of nonwoven textile do not allow water retention and
infecting due to long retention in the fibre system.
19

Fig. 8: The comparison of contact angle values.
3.2.4. Thermogravimetric analysis
It is well known that a relatively high content of phosphorus and nitrogen in materials leads to their increased
thermal stability [47].
Phosphazene compounds, containing skeletal phosphorus and nitrogen atoms, have been studied for thermal
stability many times [47]. These compounds mostly exhibit unusual thermal properties, such as flame retardation
and self-extinguish ability [47-50]. There are more possible explanations:
•
the thermal decomposition of the phosphazene-based polymers is an endothermic process when the heat of
combustion is consumed.
•
their thermal decomposition leads to the formation of PN nitrides and nonvolatile protective film formed
cover the surface of the polymer. This layer isolates the rest of the polymeric material from air oxygen –
meaning that the supply of oxygen is cut off [51-52]. If cyclo-triphosphazenes are incorporated into the
network of thermoset polymers, or into the chain structure of thermoplastics, they can increase the thermal
and flame resistance of the polymers [53].
Besides, it is assumed that the combination of the PN backbone and, fluorine containing side groups (e.g.
trifluoroethoxy group in the additives 3 and 4) can affect the thermal behaviour of the polymeric nanofibers.
Therefore, TGA and DCS analysis were carried out to determine the thermal properties of nanofibers.
20

The minimum thermal decomposition temperature (Tda) of 3 and 4 was determined from the TGA curves
(see Figs. 9, 10).
The results of the TGA analysis confirmed that the decomposition of both 3 and 4 begins below 100°C. It
follows from the course of TGA curves that 3 sublimates (Fig. 9) while 4 decomposes (Fig. 10). A
trifluoroethanole molecule, as the first decomposition product, is released from the side trifluoroethoxy groups
bound directly to the phosphorus atom of the PN ring [54] with consequent inhibition of oxygen access and
retarding effect on the combustion process.
Anyway, the remains of 4, after the thermal decomposition, can be assumed to be an unspecified form of
phosphorus nitride. This substance is highly thermally stable and could influence the thermal stability of the
materials containing it in a combination with organic polymers.
The TGA curves were determined also for Fibers A, 10B, and 10C with the aim to compare the
minimum thermal decomposition temperatures (Tda) of these Fibers (A, 10B, and 10C). It is very difficult to
describe the decomposition process, because all components (i.e. 3 or 4 and commercial polymer) decompose
independently at the same time. The Tda of fibers is highly influenced by the Tda value of the commercial
polymer.
Fig. 9: TGA curve of 3.
21

Fig. 10: TGA curve of 4.
DSC study was carried out with the additive 3 as a representative of more hydrophobic material and so
potentially better used material. As a representative fibrous material were chosen nonwoven textile spun from
mixtures PA6, PA6+3, SAN, SAN+3, because of their homogeny structure. As the example of DSC sample
spectra is presented PA6 polymer and combination with 3 additive, printed in Fig.11, the results are listed in
Tab.6.
From the results, it can be seen, that the additive 3 acts as a plasticizer, decreasing the glass transition
temperature for samples containing polymer SAN and the melt point for samples containing PA6 polymer. There
is probably a decrease in intermolecular cohesion on one side, on the other side there is an increase in chain
flexibility [55, 56], which have an effect on mechanical properties (eg. on elasticity modulus). The results also
show that the presence of additive 3 positively influenced the degradation of polymeric material themselves. In
both cases the start degradation temperature increased, in case of samples containing PA6 by 4.1°C, in case of
SAN samples by 42.6°C.
22

Tg
PA6
0
-1
-2
-3
-4
-5
-6
PA6+3
Tm
Tm
exo down
100
200
300
400
500
Temperature (°C)
Fig. 11: DSC analysis of pure PA6 polymer and combination with 3 additive.
Table 6: Results of samples from DSC analysis.
Sample Tg (°C)
Tm (°C)
Start of degradation (°C)
PA6
PA6+3
SAN
46.9
46.3
220.1
397.0
401.1
325.7
368.3
218.0
100.2
98.7
-
-
SAN+3
4. Conclusion
The phosphazene derivatives hexakis(2,2,2–trifluoroethoxy)-cyclo-triphosphazene (3) and tetrakis (2,2,2–
trifluoroethoxy)-diamido-cyclo-triphosphazene (4) were synthesized, characterized, and used as the additives in
some commercial polymers (PA6, PS, SAN, ASA, PESU) to form more hydrophobic nanofiber materials created
by the electrospinning technology.
Commonly, fibrous materials themselves exhibit a low wettability in comparison with the smooth surface of
a compact polymer matrix. In our case, it was found out that also the presence of both additives (3 or 4) in
polymers positively increase the hydrophobicity of organic polymeric fibers, whereas major impact on the
resulting value of the water contact angle in Fibers A, B, C has the original hydrophobic character of a
23

commercial polymer itself. More significant changes in the hydrophobic behaviour were observed in the
originally “hydrophilic” polymers (PA6, ASA, SAN) as opposed to the originally “hydrophobic” polymers (PS
and PESU).
It was found out that increased number of -OCH2CF3 groups, bound in the additive 3 or 4, leads to a higher
hydrophobicity of the resulting nanofibers. Generally, the presence of 3 or 4 unambiguously improves the final
hydrophobic properties of the whole nonwoven material.
The minimum thermal decomposition temperatures of additives 3 and 4 were determined with the aim to
describe their thermal behaviour in the Fibers B and C, too. The predominant presence of commercial organic
polymer in fibers complicates significantly the explaining the course of TGA curves and, subsequently,
describing the role of phosphazene additives as fire suppressants. We have confirmed the theoretical assumption
by DSC, that the additive acts as a plasticizer which has the potential to increase the elasticity of the modified
material. These results, especially increasing material hydrophobicity, with the combination of the fact that it is
light porous nonwoven textile, promise to potential use as protective covering material, eg. for cultural heritage
objects in depositary.
5. Acknowledgements
The authors are highly thankful to X-ray Diffraction and Bio-SAXS Core Facility of CEITEC for experimental
support, FR-TI1/413, Ministry of Industry & Trade and the SEM analysis was performed by D. Pavlinak.
6. References
1. T. Kmecko, X. Wang, and P. Wisian-Neilson. J Inorg Organomet Polymer Mater, 17(2), 413-421 (2007).
2. A. Konig, and S.E. Basf. U.S. Patent, 2012152805 (2012).
3. S. Zhang, N.A. Rind, N. Tang, H. Liu, X. Yin, J. Yu, and B. Ding. William Andrew Publishing. 365-389
(2019)
4. S. Zhang, H. Liu, J. Yu, W. Luo, and B Ding. Journal of Materials Chemistry A, 4(16), 6149-6157 (2016).
5. J.P. Feng, J. Wang, W.T. Hwang, and Y.M. Jo. Journal of Applied Polymer Science, 48166 (2019).
6. S.X. Wang, C.C. Yap, J. He, C. Chen, S.Y. Wong, and X. Li. Nanotechnology Reviews, 5(1), 51-73 (2016).
7. X. Wang, and B.S. Hsiao. Current opinion in chemical engineering, 12, 62-81 (2016).
8. P. Nikoloudis, R. Lausch, and T. Loewe. U.S. Patent, 9539527 (2017).
9. A. Hauk, E. Jurkiewicz, I. Pahl, T. Loewe, and R. Menzel. European Journal of Pharmaceutical Sciences, 120,
191-198 (2018).
24

10. M.G. Hajra, K. Mehta, and G.G. Chase. Separation and purification technology, 30(1), 79-88 (2003).
11. D. Groitzsch, and E. Fahrbach, U.S. Patent 4618524 (1986).
12. H. Xu, J. Liu, X. Li, C. Zhang, and Y. Wang. Separation Science and Technology, 51(1), 41-48 (2016).
13. S.R. Skaggs, J. Kaizerman, and R.E. Tapscott. Phosphorus Nitrides As Fire Extinguishing Agents. In:
National institute of standards and technology. http://www.nist.gov/el/fire_research/upload/R0000235.pdf
14. R. Voznicova Koci, J. Taraba, J. Prihoda, and M. Alberti. Polyhedron. 27(9-10), 2077-2082 (2008).
15. S. Ramakrishna. An introduction to electrospinning and nanofibers. 1. Hackensack, NJ: World Scientific,
c2005, xi, 382 p. ISBN 9812564543.
16. A. Frenot, and I. S. Chronakis. Curr Opin Colloid, 8(1), 64-75 (2003).
17. J. Doshi, and D.H. Reneker. J Electrostat, 35(2-3), 151-160 (1995).
18. C. Yue, J. Liu, H. Zhang, L. Dai, B. Wei, and Q. Chang. Heliyon, 4(9), e00809 (2018).
19. P. Liu, L. Niu, X. Tao, X. Li, Z. Zhang, and L. Yu. Applied Surface Science, 447, 656-663 (2018).
20. H.R. Allccock, R. Harry, J.S. Rutt, and R.J. Fitzpatrick. Chemistry of Materials, 3(3), 442-449 (1991).
21. http://www.archive.org/stream/americanchemical171895balt#page/279/mode/1up
22. J. Liebig, and F. Wohler. Nachtrag der Redaction. Annalen der Pharmacie, 11(2), 139-150 (1834).
23. P. Wisian-Neilson, H.R. Allcock, and K.J. Wynne. Inorganic and organometallic polymers II: advanced
materials and intermediates: developed from a symposium sponsored by the Division of Polymer Chemistry,
Inc., at the 205th National Meeting of the American Chemical Society, Denver, Colorado, March 28-April 2,
1993. 1. Washington, DC: ACS, 1994. ISBN 0841230366.
24. J.H. Gladstone, and J.D. Holmes. J Chem Soc, (17), 225-237 (1864).
25. J.K Fincham, M.B Hursthouse, H.G. Parkes, L.S. Shaw (Nee Gozen), and R.S. Shaw. J. Chem. Soc., Chem
Commun, (5), 252-254 (1985).
26. H.R. Allcock. Recent advances in phosphazene (phosphonitrilic) chemistry. DOI: 10.1021/cr60278a002.
ISBN 10.1021/cr60278a002. http://pubs.acs.org/doi/abs/10.1021/cr60278a002
27. R. Schenck, and G. Romer. Berichte der deutschen chemischen Gesellschaft (A and B Series), 57(8), 1343-
1355 (1924).
28. J.L. Schmutz, and H.R. Allcock. Inorganic Chemistry, 14(10), 2433-2438 (1975).
29. H.R. Allcock. Phosphorus-nitrogen compounds; cyclic, linear, and high polymeric systems. New York:
Academic Press, 1972. ISBN 0120505606.
30. M. Alberti. Diploma thesis, Univerzita J. E. Purkyně (Masarykova univerzita) (1983).
25

31. R. Ratz, H. Schroeder, H. Ulrich, E. Kober, and C.H. Grundmann. J Am Chem Soc, 84(4), 551-555 (1962).
32. Z. Tian, C.H. Chen, H.R. and Allcock. Macromolecules, 48(5), 1483-1492 (2015).
33. W. Armarego. Purification of Laboratory Chemicals. 4th Ed. Oxford: Butterworth Heinemann, 1996, xi, 529
s. ISBN 07-506-3761-7.
34. R. Bacovska, P. Wisian-Neilson, M. Alberti, J. Prihoda, L. Zarybnicka, and Z. Vorac. React Funct Polym,
100, 53-63 (2016).
35. CrystalClear: Rigaku (2014). CrystalClear-SM expert 2.1 b32
36. CrysAlisPro: Agilent Technologies (2013). CrysAlisPro Software system, Version 1.171.37.35
37. G.M. Sheldrick. A short history of SHELX. Acta Crystallogr A Found Crystallogr, 64(1), 112-122 (2008).
38. C.A. Schneider, W.S. Rasband, and K.W. Eliceiri. NIH Image to ImageJ: 25 years of image analysis. Nature
Methods, 9(7), 671-675 (2012).
39. Advex Instruments - See system [online]. Advex Instruments, s. r. o., 2008. http://www.advex-instruments.cz
40. M. Żenkiewicz. Journal of Achievements in Materials and Manufacturing Engineering. International OCSCO
World Press, 2007, 2007(Vol. 24, nr 1), 137-145.
41. ČSN EN 828. ČSN EN 828: Adhesives – Wettability – Determination by measurement of contact angle and
surface free energy of solid surface. Praha: Úřad pro technickou normalizaci, metrologii a státní zkušebnictví,
2013.
42. L. Daasch, and D. Smith. Analytical Chemistry, 23(6), 853–868 (1951).
43. X. Zong, K. Kim, D. Fang, S. Ran, B.S. Hsiao, and B. Chu. Polymer, 43(16), 4403-4412 (2002).
44. P.J. Brown, and K. Stevens. “Nanofibers and nanotechnology in textiles”. Cambridge: Woodhead Publishing,
in association with The Textile Institute, 2007. ISBN 9781845691059.
45. E. Kostalova, E. Zemanova, P. Mikes, J. Soukupova, H. Matheisova, and K. Klouda. Electrospinning and
electrospraying of polymer solution with spherical fullerenes. In: NANOCON 2012, 4th International
Conference Proceedings. 1. Ostrava: TANGER Ltd, 2012, s. 435-440. ISBN 978-80-87294-32- 1. http://konsys-
t.tanger.cz/files/proceedings/04/reports/630.pdf
46. M.G. Mckee, C.L. Elkins, and T.E. Long. Polymer, 45(26), 8705-8715 (2004).
47. R. Liu, and W. Wang. Polym Degrad Stab, 94(4), 617-624 (2009).
48. R. De Jaeger. Prog Polym Sci, 23(2), 179-276 (1998).
49. C.H.W. Allen. Chem Rev, 91(2), 119-135 (1991).
26

50. G.F. Levchik, Y.G. Grigoriev, A.I. Balabanovich, and M. Klatt. Polymer International: Special Issue: Fire
Retardant Polymers. Society of Chemical Industry, 49(10), 1095-1110 (2000).
51. J.W. Gu, G.Ch. Zhang, S.L. Dong, Q.Y. Zhang, and J. Kong. Surf Coat Technol, 201(18), 7835-7841 (2007).
52. S.Y. Lu, I. Hamerton. Prog Polym Sci, 27(8), 1661-1712 (2002).
53. H.R. Allcock, R.L. Kugel, and K.J. Valan. Inorg Chem, 5(10), 1709-1715 (1966).
54. L.R. Rudnick. “Synthetics, mineral oils, and bio-based lubricants: chemistry and technology”. 2nd ed. Boca
Raton: CRC Press, c2013, xvi, 992 s. Chemical industries. ISBN 9781439855379.
55. Košková, B.: Struktura a vlastnosti vláken (skriptum). TU Liberec, 1989.
56. Kroschwitz, J.I. : Polymers: High performance polymers and composites. John Wiley&Sons Inc., 1991,
ISBN 0-471-54366-7.
27

HIGHLIGHTS
• synthesis of two phosphazene derivatives, namely hexakis(2,2,2–trifluoroethoxy)-cyclo-
triphosphazene and tetrakis(2,2,2-trifluoroethoxy)-diamino-cyclo-triphosphazene
• the use of novel compounds as additives for media filter
• improving surface properties of polymer nanolayers
• improving properties hydrofobicity