Improvement in Char Strength with an Open Cage ... - MDPI

materials Article

Improvement in Char Strength with an Open Cage Silsesquioxane Flame Retardant Yolanda Bautista *, Ana Gozalbo, Sergio Mestre and Vicente Sanz Instituto Universitario de Tecnología Cerámica, Universitat Jaume I, 12006 Castellón, Spain; [email protected] (A.G.); [email protected] (S.M.); [email protected] (V.S.) * Correspondence: [email protected]; Tel.: +34-608-066-579 Academic Editor: Thomas Fiedler Received: 27 April 2017; Accepted: 18 May 2017; Published: 23 May 2017

Abstract: Different characterization techniques were used to study the hydrolysis and condensation reaction kinetics of 3-methacryloxypropyltrimethoxysilane (MAPTMS) to obtain open cage silsesquioxane oligomers. The formation of hydrogen bonds, which condition the chemical structures of the resulting products, was identified. Improved thermal and fire resistant behavior of unsaturated polyester (UP) composites prepared with aluminium trihydroxide (ATH) and the synthesized oligomer were registered. Opened silsesquioxane structures also showed an improvement in the mechanical properties of the char formed after firing. Keywords: 3-methacryloxypropyltrimethoxysilane (MAPTMS); open silsesquioxane; hydrogen bond; fire retardance; thermal degradation

1. Introduction Unsaturated polyester (UP) resins are used in numerous applications, such as vehicle bodies, ship hulls, encapsulated electric components, and solid surface composite materials. Polyesters are flammable materials and, in the presence of the oxygen in air, the flame propagates spontaneously. In fire situations, dropping may occur while burning which may start secondary fires. To delay or prevent the fire from spreading, flame retardants are added and dispersed in polymers. The nature of these additives (halides, phosphates, hydroxides, aluminosilicates, etc.), which act by different physical and/or chemical mechanisms in the fire propagation phases, varies greatly [1,2]. Inorganic polymers such as silicones or polysiloxanes exhibit greater heat resistance than organic polymers and are widely used as lubricants, coatings, and adhesives in systems subject to high working temperatures. This greater thermal stability stems from the presence of Si–O type bonds, which have higher binding energies (about 461 kJ mol−1 ) than the C–O (358 kJ mol−1 ) and C–C (304 kJ mol−1 ) bonds in organic polymers [3]. Therefore, the incorporation of polyhedral oligomeric silsesquioxanes (POSS) into polymeric materials or composites improves mechanical properties, thermal stability, and fire resistance [4]. Much attention has been focused on the use of closed cage POSS structures in the literature, ignoring the potential of open cage structures. In the present study, organic–inorganic hybrid oligomers with silsesquioxane (SiO1.5 ) bonds in open cage structures were synthesized and polymerized together with unsaturated polyester resins and aluminium trihydroxide (ATH), a classic flame retardant. The materials obtained exhibited significantly increased thermal stability, fire resistance, and improved mechanical resistance after firing due to the open cage structures. The silsesquioxane oligomer was also structurally characterized.

Materials 2017, 10, 567; doi:10.3390/ma10060567

www.mdpi.com/journal/materials

Materials 2017, 10, 567 Materials 2017, 10, 567

2 of 14 2 of 14

2. Results Results and and Discussion Discussion 2. 2.1. Silsesquioxane Oligomers 2.1. The organic–inorganic organic–inorganic hybrid oligomers were synthesised synthesised by the the sol–gel sol–gel method method from from The 3-methacryloxypropyltrimethoxysilane (MAPTMS) which 3-methacryloxypropyltrimethoxysilane (MAPTMS) with hydrolysable methoxy groups [5], which allowed formation formation of of siloxane siloxane bonds bonds under under practically practically ambient ambient conditions conditions [6]. [6]. The The hydrolysis hydrolysis and and allowed condensation reaction kinetics depend on variables such as the H 2 O:Si mole ratio, solvent nature, pH, condensation reaction kinetics depend on variables such as the H2 O:Si mole ratio, solvent nature, temperature, and and stericsteric hindrance of theofsilanes, as well if they are catalysed in acid pH, temperature, hindrance the silanes, as as well as if they are catalysed in and acidbasic and media [6]. Oligomers with different degrees of hydrolysis and condensation can be obtained asasa basic media [6]. Oligomers with different degrees of hydrolysis and condensation can be obtained function of the different reaction conditions. a function of the different reaction conditions. In general, general,these theseoligomers, oligomers, known as POSS, unitisthat is repeated the RSiO formula In known as POSS, havehave a unita that repeated with thewith formula 3/2 . RSiO 3/2 . The term ‘silsesqui’ refers to the Si:O atom ratio, which is 1:1.5. A silsesquioxane type The term ‘silsesqui’ refers to the Si:O atom ratio, which is 1:1.5. A silsesquioxane type oligomer can oligomer can exhibit different molecularThe structures. The open may structures may becondensed, partially condensed, exhibit different molecular structures. open structures be partially whereas whereas closed arefully always fully condensed. The structures, closed structures, a prism may the closedthe ones are ones always condensed. The closed with a with prism shape,shape, may have have a triangular, square, pentagonal, or hexagonal base [7]. In POSS involving polyhedrons a triangular, square, pentagonal, or hexagonal base [7]. In POSS involving polyhedrons composed of composed siliconatoms, and oxygen atoms, the are silicon atoms are vertices located at of the polyhedron silicon and of oxygen the silicon atoms located at the of the vertices polyhedron and the edges and the edges are formed by Si–O–Si bonds. The organic substituents of the silicon trialkoxides are are formed by Si–O–Si bonds. The organic substituents of the silicon trialkoxides are oriented towards oriented towards the outside of the polyhedron, linked through the silicon vertices. the outside of the polyhedron, linked through the silicon vertices. In order order to to identify identify the the different different silicon silicon atoms, atoms, present present in in the the oligomer oligomer synthesised synthesised in in the the present present In study, the conventional Tn nomenclature proposed by Engelhardt will be used hereinafter [8]. The study, the conventional Tn nomenclature proposed by Engelhardt will be used hereinafter [8]. The letter T stands the atoms siliconwith atoms with three potential alkoxygroups. reactive groups. Tletter stands for the for silicon three potential hydroxyl hydroxyl or alkoxy or reactive The symbolThe ‘n’ symbolfor ‘n’the stands for the of other silicon atomsthe to which the first siliconto is through linked tobridging through stands number of number other silicon atoms to which first silicon is linked bridgingatoms oxygen atoms oxygen (Figure 1).(Figure 1).

Figure Scheme ofof the the nomenclature nomenclature used used to to define define the the degree of condensation of the Figure 1. 1. Scheme the trialkoxysilanes [8]. trialkoxysilanes [8].

The hydrolysis hydrolysis reaction reaction progress progress of of MAPTMS, MAPTMS, performed performed under under acid acid catalytic catalytic conditions, conditions, was was The 1H NMR spectroscopy. The protons of the MAPTMS molecule, labelled with different 1 monitored by monitored by H NMR spectroscopy. The protons of the MAPTMS molecule, labelled with different letters, and and their their signals signals in in the the spectra spectra corresponding corresponding to to the the starting starting samples samples and and after after 24 24 h h and and 48 48 hh letters, reaction, are shown in Figure 2. reaction, are shown in Figure 2. The spectrum spectrum of of the the sample sample after after the the 24 24 hh reaction reaction exhibited exhibited aa drastic drastic decrease decrease in in the the methoxy methoxy The group (OCH 3 ) signal, which was reduced to a set of signals that disappeared into the noise. This group (OCH3 ) signal, which was reduced to a set of signals that disappeared into the noise. behaviour indicates that the hydrolysis reaction took place in the first few hours of the reaction This behaviour indicates that the hydrolysis reaction took place in the first few hours of the reaction [9], [9], because in in acid acid medium medium the the hydrolysis hydrolysis rate rate is is usually usually faster faster than than the the condensation condensation rate rate [6]. [6]. In In fact, fact, in in because all the reacted samples spectra, the signal corresponding to the methoxy groups of methanol, which all the reacted samples spectra, the signal corresponding to the methoxy groups of methanol, which was generated generated as as aa by-product by-product of of the the hydrolysis hydrolysis reaction, reaction, appeared appeared at at 3.18 3.18 ppm. ppm. The The hydroxyl hydroxyl group group was (OH) bonded bonded to to the the silicon silicon after after the the hydrolysis hydrolysis reaction reaction could could correspond correspond to to the the signal signal appearing appearing at at (OH) 4.63 ppm. 4.63 ppm. In the the unreacted unreacted MAPTMS MAPTMS spectrum, spectrum, the the α, α, β, β, and and γ γ proton proton signals signals split split due due to to coupling coupling with with In vicinal protons. The hydrolysis and condensation reactions produced numerous chemical species. vicinal protons. The hydrolysis and condensation reactions produced numerous chemical species. The chemical β, β, and γ protons of these species changed slightly, producing very broad The chemical shift shiftofofthe theα,α, and γ protons of these species changed slightly, producing very peaks.peaks. broad

Materials 2017, 10, 567 Materials 2017, 10, 567

3 of 14 3 of 14

Materials 2017, 10, 567

3 of 14

Figure 2. 2. 11H NMR spectrum of 3-methacryloxypropyltrimethoxysilane (MAPTMS) at 0, 24, and 48 hh Figure Figure 2. 1H NMR spectrum of 3-methacryloxypropyltrimethoxysilane (MAPTMS) at 0, 24, and 48 h of reaction. reaction. of of reaction. 13C13 The oligomer chemical structure also by13by NMR spectroscopy (Figure 3). This The structurewas wasmonitored monitored C NMR spectroscopy Theoligomer oligomerchemical chemical structure was monitored alsoalso by C NMR spectroscopy (Figure(Figure 3). This 3). information allowed the reaction kinetics to be subsequently understood. Two major types of changes This information allowed the reaction kinetics to be subsequently understood. Two major types information allowed the reaction kinetics to be subsequently understood. Two major types of changes were observed inobserved these spectra with the reaction progress: changes inchanges the chemical shifts and in the of changes were in these spectra with the reaction progress: in the chemical shifts were observed in these spectra with the reaction progress: changes in the chemical shifts and in the duplications (Table 1). and in the duplications duplications (Table 1).(Table 1).

Figure 3.1313C NMR spectrum of MAPTMS at 0, 24, and 48 h of reaction. Figure 3. C NMR spectrum of MAPTMS at 0, 24, and 48 h of reaction. Figure 3. 13C NMR spectrum of MAPTMS at 0, 24, and 48 h of reaction.

On the one hand, the carbon signals shifted towards lower fields (left side of the spectrum). This On thevery one hand, hand,the thecarbon carbonsignals signals shifted towards lower fields side ofspectrum). the spectrum). On the one shifted towards lower fields (left(left side the This shift was significant in the case of the α carbons, which were nearest to theof reactive silicons. In This shift was very significant in the case of the α carbons, which were nearest to the reactive silicons. shift was very significant in the case of the α carbons, which were nearest to the reactive silicons. the reactions, the methoxy groups (Si–OCH3) were replaced with silsesquioxane groups (Si–O–Si) In In the reactions, themethoxy methoxy groups (Si–OCH )were were replaced with silsesquioxane groupsdensity (Si–O–Si) thewith reactions, the (Si–OCH replaced with silsesquioxane groups (Si–O–Si) lower electron affinity.groups The carbons near3)3the silicon therefore exhibited greater electron with lower electron affinity. The carbons near the silicon therefore exhibited greater electron with electron affinity. The carbons near the silicon therefore exhibited greater electron density density andlower resonated at lower fields. and resonated at lower fields. and resonated at lower fields.

Materials 2017, 10, 567

4 of 14

On the other hand, in addition to shifting, the signals of the a, c, and γ carbons (close to the carbonyl group) were duplicated in two peaks. This duplication might be due to the presence of Materials 2017, 10, 567 4 of 14 hydrogen bonds between a fraction of the methacrylate carbonyl groups (C=O) and a hydroxyl group (OH) bonded the silicon the hydrolysis Theofhydrogen bonds decreased thetocarbon On the to other hand, inafter addition to shifting,reaction. the signals the a, c, and γ carbons (close the electron density of the carbonyl group, thus resonating at lower frequencies. The intensities observed carbonyl group) were duplicated in two peaks. This duplication might be due to the presence of in the duplicated indicated that group fraction with (C=O) hydrogen wasgroup 0.35. hydrogen bondspeaks between a fraction ofthe the carbonyl methacrylate carbonyl groups and abonds hydroxyl (OH) bonded to the silicon after the hydrolysis reaction. The hydrogen bonds decreased the carbon Table 1. The most significant and duplications the 13frequencies. C NMR chemical shifts at three the electron density of the carbonylchanges group, thus resonating atoflower The intensities observed reaction times. in the duplicated peaks indicated that the carbonyl group fraction with hydrogen bonds was 0.35. 13C NMR chemical shifts at three the Reaction Table 1. The most significant changes and duplications of theTime reaction times. t=0h t = 24 h t = 48 h Reference Chemical Chemical Chemical Intensity Reaction Intensity Time t = 0Shifts h t = 24 h t = 48 h (ppm) Shifts (ppm) Ratio Shifts (ppm) Ratio

Reference

α α γ

γ a

a

c c

Chemical Shifts 5.07 (ppm) 5.07 66.16

Chemical Shifts 8.50 (ppm) 66.38 8.50 66.68 66.38 66.16 125.31 66.68 124.64 125.31 125.44 124.64 125.44 167.43 166.85 167.43 167.67 166.85 167.67

Intensity — Ratio 0.40 --0.60 0.40 0.60 0.32 0.32 0.68 0.68 0.32 0.32 0.68 0.68

Chemical Shifts 8.50 (ppm) 66.36 8.50 66.65 66.36 125.29 66.65 125.42 125.29 167.34 125.42 167.61 167.34 167.61

Intensity — Ratio 0.35 --0.65 0.35 0.34 0.65 0.66 0.34 0.34 0.66 0.66 0.34 0.66

The literature reports be used usedto toassess assessthe thepresence presence hydrogen The literature reportsthat thatinfrared infraredspectroscopy spectroscopy can be ofof hydrogen bonds in in various silane methacryloxypropyltrimethoxysilane [10], bonds various silanecondensation condensationproducts, products, such such as methacryloxypropyltrimethoxysilane [10], poly(acrylicacid)/silica acid)/silica hybrid hybrid [11], [11], and and tetraethoxysilane tetraethoxysilane [12]. poly(acrylic [12]. InInthe thepresent presentstudy, study,infrared infrared spectroscopyalso alsoconfirmed confirmedthe thepresence presence of of hydrogen hydrogen bonds bonds (Figure spectroscopy (Figure 4). 4). The Thereaction reactionproduct product −1, − 1 exhibited a very broad band centred at 3460 cm corresponding to the O–H bond The exhibited a very broad band centred at 3460 cm , corresponding to the O–H vibration. bond vibration. ofof this that most ofof the groups are in in hydrogen Thebroadness broadness thisband bandindicated indicated that most thehydroxyl hydroxyl groups areparticipating participating hydrogen −11 next to the carbonyl group C=O vibration band − bonds. The appearance of a shoulder at 1703 cm bonds. The appearance of a shoulder at 1703 cm next to the carbonyl group C=O vibration band (1716 ) isusually usuallyassociated associated with with the the presence bonds lengthen −1 )−1is (1716 cmcm presence of ofhydrogen hydrogenbonds bonds[10]. [10].Hydrogen Hydrogen bonds lengthen and weaken the C=O bond, leading to absorption at lower frequencies [11,12]. and weaken the C=O bond, leading to absorption at lower frequencies [11,12].

(a)

(b)

Figure 4. FT-IR spectrum of MAPTMS and of the MAPTMS hydrolysis and condensation product Figure 4. FT-IR spectrum of MAPTMS and of the MAPTMS hydrolysis and condensation product after after 264 h of reaction, (a) complete spectra, (b) details of the carbonyl signals. 264 h of reaction, (a) complete spectra, (b) details of the carbonyl signals.

The hydrogen bonds identified by 1313C NMR and FT-IR could be either intramolecular or The hydrogen bonds identified by C NMR andasFT-IR couldinbeFigure either5; intramolecular intermolecular interactions, or combinations of both, illustrated however, these or intermolecular interactions, or combinations of both, as illustrated in Figure 5; however, these techniques do not allow us to differentiate between the two types of interactions. techniques do not allow us to differentiate between the two types of interactions.

Materials 2017, 10, 567 Materials 2017, 10, 567 Materials 2017, 10, 567

5 of 14 5 of 14 5 of 14

Figure intramolecular and intermolecular hydrogen bonds. Figure5. 5.Chemical Chemicalstructure structureof ofthe the Figure 5. Chemical structure of the intramolecular intramolecular and and intermolecular intermolecular hydrogen hydrogen bonds. bonds. 29Si nucleus The condensation reaction progress was monitored by NMR spectroscopy of the 29 The condensation reaction progress was monitored spectroscopy monitored by by NMR NMR spectroscopy of of the the 29Si nucleus nucleus 29 (Figure 6). According to the literature [13], the chemical shifts of29 29Si for trialkoxysilanes are located (Figure shifts ofof SiSiforfor trialkoxysilanes areare located in (Figure 6). 6). According Accordingto tothe theliterature literature[13], [13],the thechemical chemical shifts trialkoxysilanes located 29Si signal to shift in the range from −35 to −75 ppm. Each alkoxide condensation reaction causes the 29 the range from −35 toto −75 in the range from −35 −75ppm. ppm.Each Eachalkoxide alkoxidecondensation condensationreaction reactioncauses causesthe the 29Si signal to shift by about 8–9 ppm towards higher fields in the spectrum, allowing four large areas to be identified, by about 8–9 ppm towards higher fields in the spectrum, allowing four large areas to be identified, corresponding to the different degrees of condensation; T0, T1, T2, and T3 (Figure 1). corresponding to the the different different degrees degrees of of condensation; condensation; T0, T0, T1, T1,T2, T2,and andT3 T3(Figure (Figure1). 1).

264 hours 264 hours

168 hours 168 hours 96 hours 96 hours

72 hours 72 hours 48 hours 48 hours 24 hours 24 hours 0 hours 0 hours -30 -30

-40 -40

-50 -50

-60 -70 ppm -60 -70 ppm Figure 6. 2929Si NMR spectra of MAPTMS at different reaction times. Figure 6. Si NMR spectra of MAPTMS at different reaction times. Figure 6. 29 Si NMR spectra of MAPTMS at different reaction times.

-80 -80

The replacement of an alkoxide group (OR) with a hydroxyl group (OH) modifies the chemical The replacement of an alkoxide group (OR) with a hydroxyl group (OH) modifies the chemical 29Si between 0.4 and 0.8 ppm [13]. The progress of the hydrolysis and condensation reactions shift of replacement an alkoxide group (OR) with aof hydroxyl group (OH) modifies the reactions chemical shift The of 29Si between 0.4ofand 0.8 ppm [13]. The progress the hydrolysis and condensation 29 the increases number0.4 of and combinations of the silicon atom chemical environments, which is reflected shift of Si between 0.8 ppm [13]. The progress of the hydrolysis and condensation reactions increases the number of combinations of the silicon atom chemical environments, which is reflected in a complex of signals in each of the four atom areas of the spectrum [14]. which is reflected in increases the structure number of the silicon environments, in a complex structureofofcombinations signals in each of the four areaschemical of the spectrum [14]. It was structure taken into that theofparticipation ofofthe atom[14]. in small cycles reduces the a complex ofaccount signals in each the four areas thesilicon spectrum It was taken into account that the participation of the silicon atom in small cycles reduces the angleItofwas the Si–O–Si bond withthat relation to the linear of species, inducing signal shift towards lower into account the participation the silicon atomaain small cycles reduces the angle of thetaken Si–O–Si bond with relation to the linear species, inducing signal shift towards lower fields (left of the spectrum). That is, in the T2 signal area, the silicons included in cycles are shifted to angle of the bond with the linear species, inducing a signal towards lower fields (left ofSi–O–Si the spectrum). Thatrelation is, in thetoT2 signal area, the silicons included in shift cycles are shifted to the left [15].of the spectrum). That is, in the T2 signal area, the silicons included in cycles are shifted to fields the left(left [15]. The[15]. data reported in the literature enabled the groups of signals in the 29Si NMR spectrum, the left The data reported in the literature enabled the groups of signals in the 29Si NMR spectrum, 29 Si NMR detailed Table 2, to be in identified. The subscripts usedgroups indicate number of hydroxyl groups still Thein data reported the literature enabled the of the signals in the spectrum, detailed in Table 2, to be identified. The subscripts used indicate the number of hydroxyl groups still present, while the2,reference in brackets indicates the linear (l), branched (r), of or hydroxyl cyclical (c) structure detailed in Table to be identified. The subscripts used indicate the number groups still present, while the reference in brackets indicates the linear (l), branched (r), or cyclical (c) structure and the number of reference silicon units forming the structure. present, while the in brackets the linear (l), branched (r), or cyclical (c) structure and the number of silicon units forming indicates the structure. and the number of silicon units forming the structure. Table 2. Band assignment in the 2929Si NMR spectra of the MAPTMS reaction. Table 2. Band assignment in the Si NMR spectra of the MAPTMS reaction.

Species Species

Chemical Band Shift (ppm) Chemical Band Shift (ppm)

Materials 2017, 10, 567

6 of 14

Table 2. Band assignment in the 29 Si NMR spectra of the MAPTMS reaction. Materials 2017, 10, 567

Species T1 T21 (3c) T1 T2 0 (3c) T2 1(3c) T21 (4c) T2 0(3c) T20 (4c) T2 1(4c) T2(5c+6c+l) T3(3c) T20(4c) T3(4c) T2(5c+6c+l) T3(r)

Chemical Band Shift (ppm)

6 of 14

−47.4 to −49.3 − 52.7 to 54.2 −47.4 to−−49.3 − 54.2 to 55.4 −52.7 to−−54.2 −55.4 to −56.1 −54.2 to−−55.4 − 56.1 to 57.3 −55.4 to−−56.1 −57.3 to 58.1 −−56.1 58.1 to 59.2 to − −57.3 − 59.2 to 60.7 −57.3 to−−58.1 −62 to −69

T3(3c) − 58.1 to −59.2 T3(4c) −59.2 to −60.7 T3(r) −69plotted versus reaction time in Figure 7. The integrals of the signals of silicon T0, T1, T2, and−62 T3toare

In the firstThe 24 integrals h, the highly reactive silicons were completely consumed, forming structures of the signals of T0 silicon T0, T1, T2, and T3 are plotted versus reaction time in Figure with a greater The T1 rapidly, exhibiting a maximum concentration 7. In degree the firstof 24condensation. h, the highly reactive T0silicon siliconsformed were completely consumed, forming structures with at 24 ah greater of the reaction. Their concentration then progressively to condensation, giving degree of condensation. The T1 silicon formed rapidly,decreased exhibiting adue maximum concentration 24 h of the concentration progressively due tothe condensation, givingof the rise toatT2 and T3 reaction. silicon. Their A large number ofthen T2 silicons weredecreased present after first few hours rise with to T2 aand T3 silicon. number of T2 Their silicons were present then after decreased the first few hours of reaction, maximum at A 24large h of the reaction. concentration slightly asthe a result reaction, with a maximum at 24 h of the reaction. Their concentration then decreased slightly a of the balance between the newly formed T2 silicons and those that were transformed into T3assilicons. result of the balance between the newly formed T2 silicons and those that were transformed into T3 The T3 silicon concentration increased continuously due to the reaction of less condensed silicons. silicons. The T3 silicon concentration increased continuously due to the reaction of less condensed Finally, under the condensation conditions used, a distribution of T2 and T3 silicons was attained that silicons. Finally, under the condensation conditions used, a distribution of T2 and T3 silicons was depended on the final structures that formed. attained that depended on the final structures that formed. The reactivity of theofsilicon species generally decreased as their of condensation increased, The reactivity the silicon species generally decreased as degree their degree of condensation as had already as been by other researchers [16,17]. [16,17]. increased, hadobserved already been observed by other researchers

Figure 7. Relative concentration of the types of silicon atoms sol–gelreaction reactionproducts productswith with time. Figure 7. Relative concentration of the types of silicon atoms in in thethe sol–gel time. Values of the integrals obtained29in the 29Si NMR spectrum. Values of the integrals obtained in the Si NMR spectrum.

Further analysis of T2 silicons (Figure 8) revealed that about half of these participated in tetramer

Further analysis of T2 group silicons 8) revealed that about remaining half of these participated in tetramer cycles with a hydroxyl (T2(Figure 1(4c)), with their concentration constant over time. The T2 silicons that part of the pentamer and hexamer cycles and linear species cyclessum withofa the hydroxyl group (T2formed (4c)), with their concentration remaining constant over time. The sum 1 (T2(5c+6c+l)) was of the same order, though it decreased with time. The decrease might be due to the of the T2 silicons that formed part of the pentamer and hexamer cycles and linear species (T2(5c+6c+l)) of the linearitspecies into cyclical species. transformation of thetolinear species in was oftransformation the same order, though decreased with time. The This decrease might be due the transformation acid medium has been reported in the condensation reactions of other silanes, such as of the linear species into cyclical species. This transformation of the linear species in acid medium methyltrimethoxysilane [16], methyltriethoxysilane [15], and tetraethoxysilane [18]. The presence of has been reported in the condensation reactions of other silanes, such as methyltrimethoxysilane [16], T2 silicons in trimer cycles was very small. methyltriethoxysilane [15], and tetraethoxysilane [18]. The presence of T2 silicons in trimer cycles was The stability of T21(4c) silicons might be explained by the existence of hydrogen bonds, described very small. above, which could inhibit the nucleophilic attack of the condensation reaction. The T21(4c) silicon

Materials 2017, 10, 567

7 of 14

fraction identified with respect to the total existing silicons was about 0.30. This value was very similar to that of the carbonyl group fraction affected by hydrogen bonds, identified by 13C NMR, Materials 2017, 10, 567 7 of 14 which reinforces the role of these bonds in hindering silicon atom reactivity.

Materials 2017, 10, 567

7 of 14

fraction identified with respect to the total existing silicons was about 0.30. This value was very similar to that of the carbonyl group fraction affected by hydrogen bonds, identified by 13C NMR, which reinforces the role of these bonds in hindering silicon atom reactivity.

Figure Evolutionwith withtime time of of the the fraction fraction of inin the sol–gel reaction products. Figure 8. 8.Evolution ofT2 T2type typesilicon siliconatoms atoms the sol–gel reaction products. 29Si NMR spectrum. 29 Values of the integrals obtained in the Values of the integrals obtained in the NMR spectrum.

Analysis of the T3 silicons (Figure 9) revealed a minor presence of prismatic structures (T3(4c) The stability of T21 (4c) silicons might be explained by the existence of hydrogen bonds, described and T3(3c)). Most of the fully condensed silicons formed part of the branched structures, T3(r), less above, which could inhibit the nucleophilic attack of the condensation reaction. The T21 (4c) silicon subject to the stresses of several cycles. These results were observed, with minor variations, for the fraction identified to the total existing silicons was about 0.30. This value was very similar different batcheswith of therespect oligomer prepared. 8. Evolution with time of the fraction of T2 silicon atoms in theidentified sol–gel reaction products. to that ofFigure the carbonyl group fraction affected bytype hydrogen bonds, by 13 C NMR, which Values of the integrals obtained in the 29Si NMR spectrum. reinforces the role of these bonds in hindering silicon atom reactivity. Analysis of the T3 T3 silicons (Figure minorpresence presence prismatic structures (T3(4c) Analysis of the silicons (Figure9)9)revealed revealed aa minor of of prismatic structures (T3(4c) and T3(3c)). Most of the fully condensed silicons formed part of the branched structures, T3(r), and T3(3c)). Most of the fully condensed silicons formed part of the branched structures, T3(r), less less subject to the of several cycles. wereobserved, observed, with minor variations, for the subject to stresses the stresses of several cycles.These These results results were with minor variations, for the different batches of the oligomer prepared. different batches of the oligomer prepared.

Figure 9. Evolution with time of the fraction of T3 silicon atoms in the sol–gel reaction products. Values of the integrals obtained in the 29Si NMR spectrum.

Hydrolysis and condensation in the sol–gel process give rise to numerous species, generically known as silsesquioxanes, having the general formula: Tn(OH)x(OCH3)y, where T = RSiO1.5-m/2n, with (mFigure = x +Figure y), to the nomenclature used by Eisenberg et Thereaction molecular masses of the 9. Evolution with time of fraction the fraction of T3 silicon atoms in[19]. the sol–gel reaction products. 9. according Evolution with time of the of T3 silicon atoms in al. the sol–gel products. Values 29 29 chemical species arising as a result of MAPTMS hydrolysis and condensation were quantified by of the integralsinobtained the Si NMR spectrum. of theValues integrals obtained the SiinNMR spectrum. mass spectrometry. Data analysis then allowed the main chemical structures to be identified. Hydrolysis and condensation in the sol–gel process give rise to numerous species, generically

Hydrolysis and condensation the sol–gel process givex(OCH rise to3)ynumerous known as silsesquioxanes, havingin the general formula: Tn(OH) , where T = species, RSiO1.5-m/2ngenerically , with known having the general Tn (OH) (OCH = RSiO y , where T 3 )molecular 1.5-m/2n , (m as = xsilsesquioxanes, + y), according to the nomenclature usedformula: by Eisenberg et al. x[19]. The masses of the with (m = x + y), according nomenclature used by Eisenberg et al. [19]. The molecular masses chemical species arisingto asthe a result of MAPTMS hydrolysis and condensation were quantified by of mass spectrometry. Data analysis then allowed the main chemical structures to be identified. the chemical species arising as a result of MAPTMS hydrolysis and condensation were quantified by mass spectrometry. Data analysis then allowed the main chemical structures to be identified.

Materials 2017, 10, 567

8 of 14

Materials 2017, 10, 567 Materials 2017, 10, 567

8 of 14 8 of 14

The mass spectra reaction(Figures (Figures10 10and and11, 11,respectively) respectively) displayed The mass spectraafter after24 24hhand and264 264 hh of of the the reaction displayed Thedistribution mass spectraofafter 24 h andcorresponding 264 h of the reaction (Figures 10 and 11, respectively) displayed a primary the species, to the number of bonded monomers (n = a primary distribution of the species, corresponding to the number of bonded monomers 2(ndimers; = 2 a 3primary distribution of theetc.). species, corresponding to the of bonded monomers (n = 2 n =dimers; trimers; 4 tetramers, At the sameAt value n, a number secondary n = n3=trimers; n = 4 tetramers, etc.). the of same value of n, distribution a secondary corresponding distribution dimers; n = 3oftrimers; n = 4 tetramers, etc.).the Atvalue the same value ofvalue n, a of secondary distribution to corresponding the variation modification of both x and y was observed. In the tom, theby variation of m, by modification of of both the the value of x and the value of y was corresponding to the variation of m, by modification of both the value of x and the value of y was secondary distribution, the value of m indicated the number of intramolecular cycles, while the value observed. In the secondary distribution, the value of m indicated the number of intramolecular cycles, observed. In the secondary distribution, the value of m indicated the number of intramolecular cycles, the value y indicated the presence ofalkoxide unhydrolyzed alkoxide groups. of while y indicated the of presence of unhydrolyzed groups. while the value of y indicated the presence of unhydrolyzed alkoxide groups.

Figure 10. Electrospray ionization time-of-flight mass spectrum (ESI–TOF MS) of the MAPTMS Figure 10.10.Electrospray mass spectrum spectrum(ESI–TOF (ESI–TOFMS) MS)ofofthe theMAPTMS MAPTMS Figure Electrospray ionization ionization time-of-flight time-of-flight mass hydrolysis and condensation product after 24 h of reaction. hydrolysis and condensation product after 24 h of reaction. hydrolysis and condensation product after 24 h of reaction. muestra MEMO 13 dias scan pos MEOH 20 V T5 muestra MEMO 13 dias pos 20 V SCIC_RG_005 1 (1.111) Smscan Sm (Mn, 2x0.50) Scan ES+ T MEOH T(Mn, 2x0.90); 6

5

SCIC_RG_005 1 (1.111) Sm (Mn, 2x0.90); Sm (Mn, 2x0.50) T6 100 100 1116.3 1116.3 6119917 946.5 6119917 T7 946.5 5138915 T7 5138915 1134.5 1134.5 3228595 3228595 776.2 1304.7 776.2 1304.7 3016128 3626665 3016128 3626665

Scan ES+ 3.23e6 3.23e6

%%

T4 T4

T6 T6

T7 T7

578.5 578.5 606631 606631

0 0

600 600

1474.7 1474.7 972982 972982

977.8 977.8 820089 820089

800 800

1000 1000

T8 T8

1200 1200

1400 1400

1600 1600

1800 1800

2000 2000

m/z m/z

Figure 11. ESI–TOF MS of the MAPTMS hydrolysis and condensation product after 264 h of reaction. Figure ESI–TOFMS MSofofthe theMAPTMS MAPTMS hydrolysis hydrolysis and 264 h of reaction. Figure 11.11. ESI–TOF and condensation condensationproduct productafter after 264 h of reaction.

Mass spectroscopy showed that, after 24 h of reaction, no independent silane molecules Mass spectroscopy showed that, after 24 h of reaction, no independent silane molecules (monomers T1) remained. The existing consisted of 2 tono 7 silsesquioxane Table Mass spectroscopy showed that, oligomers after 24 h of reaction, independentmolecules. silane molecules (monomers T1) remained. The existing oligomers consisted of 2 to 7 silsesquioxane molecules. Table 3 lists the major species, corresponding to about 68% by weight of the sample. They were mainly (monomers T1) remained. The existing oligomers consisted of 2 to 7 silsesquioxane molecules. Table 3 3 lists the major species, corresponding to about 68% by weight of the sample. They were mainly linear dimer species, and trimer structures, into addition to single-cycle Although species lists the major corresponding about 68% by weighttetramers. of the sample. Theymost werewere mainly linear linear dimer and trimer structures, in addition to single-cycle tetramers. Although most were species with fully hydrolysed alkoxide groups, some oligomers still had at least one unhydrolyzed alkoxide dimer structures, in addition to single-cycle Although most were species with withand fullytrimer hydrolysed alkoxide groups, some oligomers tetramers. still had at least one unhydrolyzed alkoxide group. Table 3 includes agroups, schematic illustration of the possible structures for each mass/charge ratio: fully hydrolysed alkoxide some oligomers still had at least one unhydrolyzed alkoxide group. group. Table 3 includes a schematic illustration of the possible structures for each mass/charge ratio: the black dots represent the silicon atoms, while the segments between two black dots represent the Table includes a schematic illustration the possible structures fortwo each mass/charge ratio: the 3 black dots represent the silicon atoms,ofwhile the segments between black dots represent thethe links between silicon atoms through oxygen atoms, i.e., Si–O–Si type bonds. The species with an black represent theatoms silicon atoms,oxygen while atoms, the segments between black represent linksdots between silicon through i.e., Si–O–Si type two bonds. Thedots species with anthe unhydrolyzed group have not been included for the sake of simplicity. However, these would be links between silicon through oxygen atoms, type bonds. Thethese species with unhydrolyzed groupatoms have not been included for the i.e., sakeSi–O–Si of simplicity. However, would be an similar to those illustrated, with one of the OH groups being replaced with an OCH3 group. unhydrolyzed group have not been included for the sake of simplicity. However, these would be similar to those illustrated, with one of the OH groups being replaced with an OCH3 group. similar to those illustrated, with one of the OH groups being replaced with an OCH3 group.

Materials 2017, 10, 567

9 of 14

Materials2017, 2017,10, 10,567 567 Materials 2017, 10, 567 Materials

of14 14 of 14 999of

Table 3. 3. Summary Summary of the the major species species detected inoligomer theoligomer oligomer by mass mass spectrometryafter after 24 24 hhhhof of Table 3. Summary of the major species detected in the oligomer by mass spectrometry after 24 of Table of major detected the spectrometry after Table 3. Summary of the major species detected in thein by by mass spectrometry 24 reaction. reaction. reaction. Materials 2017, 10, 567 9 14 of reaction.Materials 99 of of Materials 2017, 2017, 10, 10, 567 567 of 14 14

Experimental Experimental Experimental Table 3. Summary of the major species detected in the oligomer by mass spectrometry after 24 h of

Integration Species Structuresafter Integration Species Structures Integration Species Structures Table 3. Summary of detected Table of the the major major species species detected in in the the oligomer oligomer by by mass mass spectrometry spectrometry after 24 24 h h of of Experimental m/z Integration Species Structures m/zValue Value3. Summary m/z Value m/z Value reaction. reaction.

reaction. 417.1 417.1 417.1

20% 20% 20%

417.1 Experimental Experimental

(OH)444Na Na+++(l) (l) TT222(OH) (OH) Na (l) T T (OH) Na+ (l)

20%

Experimental m/z Value m/z 433.1 433.1 433.1 m/z Value Value 433.1 417.1 417.1 417.1

2

4

Integration Species Integration Species Integration Species +++ 6% (OH)333(OCH (OCH333)Na )Na+(l) (l) 6% TT222(OH) (OH) (OCH )Na (l) 6% T 6%20% T2 (OH) + (l) (l) 3 (OCH 3 )Na T 2(OH) 4Na + 20% T 20% T22(OH) (OH)44Na Na+ (l) (l)

604.8

604.8 604.8 604.8

433.1 433.1 433.1

15%6% 6%

15% 15% 15%

618.9

618.9 618.9 618.9 604.8

8% 8% 8% 8%15%

774.7

774.7 774.7 774.7 618.9

618.9 618.9

8% 11%8% 8%

(OH) Na +)Na (c) ++ (l) TTT (OH) Na (c) 444(OH) 444Na (c) T33T (OH) 4(OCH (c)+ (l) T 4(OCH 4 (OH) 4 Na333)Na T3(OH) (OH) 4(OCH )Na (l)

788.7

774.7 774.7 788.7 788.7 788.7 774.7

11% 8% 8% 8% 11% 8%

11%

T 4(OH)4Na+ (c) T 4Na + (c) (OH) (OCH )Na (c) TT4T 44(OH) (OH) (OCH )Na (c) T 333(OCH 333)Na T44(OH) (OH) 4Na (c)++++(c) (c) 4 (OH) 3 (OCH 3 )Na

788.7 788.7 788.7

8% 8% 8%

T 4(OH)3(OCH3)Na++ (c) T T44(OH) (OH)33(OCH (OCH33)Na )Na+ (c) (c)

6%

604.8 604.8

15% 15%

11% 11% 11%

Structures Structures Structures

(OH)555Na Na+++(l) (l) TT333(OH) (OH) Na (l) T

+ (l) + T3 (OH) T 2(OH) 3(OCH 5 Na3)Na + (l) T T22(OH) (OH)33(OCH (OCH33)Na )Na+ (l) (l) +++ (l) (OH) (OCH )Na (l) TT3T 33(OH) (OH) (OCH )Na T 444(OCH 333)Na + (l) +(l) T3(OH) 5Na)Na (OH) (OCH (l)

3

4 3 + (l) T 5Na T33(OH) (OH) 5Na+ (l) +++

+

After264 264hhhof ofthe thereaction, reaction,the theleast leastcondensed condensedoligomers oligomerswere werethe thetetramers tetramers(n (n===4). 4).Table Table444lists lists After 264 of the reaction, the least condensed oligomers were the tetramers (n 4). Table lists After After the 264 h of the reaction, the least condensed oligomers thesample. tetramers (nwere = 4).oligomers Table 4 the major major species, corresponding to about about 68% 68% by weight weightwere of the the sample. These were oligomers the major species, corresponding to about 68% by weight of the sample. These were oligomers species, corresponding to by of These lists the major species, corresponding to the about 68% by weight of the sample. These(n oligomers After 264 h of of the reaction, reaction, the least condensed oligomers were the tetramers (nwere 4). Table 4 lists obtained by condensation condensation of four four toleast eight silane molecules, with high number of4). intramolecular obtained by condensation of four to eight silane molecules, with high number intramolecular obtained by of to eight silane molecules, high number intramolecular After 264 condensed oligomers were tetramers ===of Table After 264 h h of the the reaction, the least condensed oligomerswith wereaaathe the tetramers (n of 4). Table 44 lists lists the and major species, corresponding toalkoxide about 68% by weight weight of the sample. These were oligomers of obtained by condensation of four to eight silane molecules, with a high number of intramolecular cycles and complete hydrolysis of the alkoxide groups. The table includes a schematic illustration of the major species, corresponding to about 68% by of the sample. These were oligomers cycles and complete hydrolysis of the alkoxide groups. The table includes a schematic illustration cycles complete hydrolysis of the groups. The table includes a schematic illustration the major species, corresponding to about 68% by weight of the sample. These were oligomers of obtained by condensation ofalkoxide four to eight eight silane molecules, with aa high high number of of intramolecular intramolecular by condensation of four to silane molecules, with number theobtained possible structures for each mass/charge ratio. cycles and the complete hydrolysis of the groups. The table includes a schematic illustration of the possible structures for each mass/charge ratio. possible structures for each mass/charge ratio. obtained by condensation of four to eight silane molecules, with a high number of intramolecular cycles and complete hydrolysis of the alkoxide groups. The table includes aa schematic illustration of cycles and complete hydrolysis of alkoxide groups. The illustration of The structural structural analysis above indicates that the oligomers oligomers consisted mainly of of four to to eight eight The structural analysis above indicates that the oligomers consisted mainly of four to eight The analysis above indicates that the consisted mainly four cycles and complete hydrolysis of the the alkoxide groups. The table table includes includes a schematic schematic illustration of the possible structures for each mass/charge ratio. the possible structures for each mass/charge ratio. the possible structures for each mass/charge ratio. silsesquioxane molecules containing unreacted hydroxyl groups (hydrolysed silanes but not yet silsesquioxane molecules containing unreacted hydroxyl groups groups (hydrolysed silanes but not yet silsesquioxane molecules containing unreacted hydroxyl (hydrolysed silanes but not yet the possible structures for each mass/charge ratio. The structural analysis above indicates that the oligomers consisted mainly of four to eight The structural analysis above indicates that the oligomers consisted mainly of four to eight The structural analysis above indicates that the oligomers consisted mainly of four to eight condensated). The participation of these groups in hydrogen bonds with the methacrylate groups The structural analysis unreacted above indicates thatin oligomers consisted mainly of four to eight condensated). Thecontaining participation of these these groups inthe hydrogen bonds with the thesilanes methacrylate groups condensated). The participation of groups hydrogen bonds with methacrylate groups silsesquioxane molecules hydroxyl groups (hydrolysed but not yet silsesquioxane molecules containing unreacted hydroxyl groups (hydrolysed silanes but not yet silsesquioxane molecules unreacted hydroxyl groups (hydrolysed silanes but not silsesquioxane molecules containing containing unreacted hydroxyl groups (hydrolysed silanes butnumber not yet yetof prevented the condensation condensation reactions from from concluding. The oligomers exhibited high number of prevented the condensation reactions from concluding. The oligomers exhibited high number of prevented the reactions concluding. The oligomers exhibited aaa high The participation of these groups in hydrogen bonds with the methacrylate groups condensated).condensated). The participation of these groups in hydrogen bonds with the methacrylate groups condensated). The participation of groups in bonds the groups condensated). Thewithout, participation of these these groups the in hydrogen hydrogen bonds with withstructures the methacrylate methacrylate groupsof cycles per molecule molecule without, however, forming the closed polyhedral polyhedral structures characteristic of cycles per molecule without, however, forming the closed polyhedral structures characteristic of cycles per however, forming closed characteristic prevented the condensation condensation reactions from concluding. concluding. The oligomers oligomers exhibiteda aahigh high number number of of prevented theprevented condensation reactions reactions from concluding. The oligomers exhibited the from The exhibited prevented the condensation reactions from concluding. The oligomers exhibited a high high number number ofof POSS. POSS. POSS. cycles per molecule without, however, forming the closed polyhedral structures characteristic of cycles per without, however, forming closed structures characteristic of cycles per molecule however, forming the closedthe polyhedral structures characteristic of POSS. cycles without, per molecule molecule without, however, forming the closed polyhedral polyhedral structures characteristic of POSS. POSS. POSS. Table4. 4.Summary Summaryof ofthe themajor majorspecies speciesdetected detectedin inthe theoligomer oligomerby bymass massspectrometry spectrometryafter after264 264hhhof of Table 4. Summary of the major species detected in the oligomer by mass spectrometry after 264 of Table Table 4. Summary of the major species detected in the oligomer by mass spectrometry after 264 h reaction. reaction. reaction. Table 4. Summary of the major species detected in the oligomer by mass spectrometry after 264 h of

of reaction.

Table Table 4. 4. Summary Summary of of the the major major species species detected detected in in the the oligomer oligomer by by mass mass spectrometry spectrometry after after 264 264 h h of of

reaction. Experimental reaction. Experimental Experimental reaction. Integration Integration Integration m/z Value m/z Value m/z Value Experimental

Experimental Integration Experimental Experimental m/z Value Integration m/z Value Integration Integration m/z m/z Value Value

1115.3 1115.3 1115.3

1115.3

1115.3 1115.3 1115.3

945.5 945.5 945.5

945.5

945.5 945.5 945.5

664.3 664.3 664.3 664.3

Species Species Species

Structures Structures Structures

Species Species Species

Species

15% 15% 15%

(OH)222Na Na+++(3c) (3c) (OH) Na (3c) TTT666(OH)

15% 15% 15%

+ (3c) T 2Na++ (3c) T6 (OH) 2 Na T66(OH) (OH) 2Na + (3c)

15%

13% 13% 13%

13% 13% 13% 13%

9%9% 9% 9%

664.3 664.3 664.3 1303.7 1303.7 1303.7 1303.7 1303.7 1303.7 1303.7

9% 9% 9%

578.5 1133.5

578.5 1133.5

9% 9%

775.2

7%

Materials 2017, 10, 567

Structures Structures Structures Structures

T6(OH)2Na (3c)

(OH)333Na Na+++(2c) (2c) (OH) Na (2c) TTT555(OH)

T55(OH) (OH)Na 3Na (2c) ++ (2c) T T5 (OH) 3 33Na T5(OH) Na+ (2c) (2c) +

(OH)333Na Na222+2+2+2(3c) (3c) (OH) Na (3c) TTT777(OH) T7(OH)3Na2+2(3c)

T 7(OH)3Na 2 +2(3c) +++2 7(OH) 3Na 2(3c) (3c) 7(OH) (OH) 3Na Na T(OH) TT (3c) (OH) 77T 7T 333Na 2+ (3c) +(3c) 7(OH) 3Na T + (3c) 7(OH)3Na (3c) T + 7 (OH) 3 Na (3c) T T7 (OH)3 Na + (3c) +2

+2 +2 (2c) T6(OH) 4Na T6 (OH) 4 Na 2 2 (2c) +2 +2 4Na T6(OH) T6 (OH) (2c) 4 Na 2 2 (2c)

T4(OH)4Na+ (c)

10 of 14

Materials 2017, 10, 567 Materials Materials 2017, 2017,10, 10,567 567 Materials 2017, 10, 567

10 of 14 10 10 of of14 14 10 of 14

Materials 2017, 10, 567

10 of 14 578.5 578.5 578.5 1133.5 578.5 1133.5 1133.5 1133.5

Experimental m/z Value

9% 9% 9% 9%

Integration

775.2

775.2 775.2 775.2 775.2

7% 7% 7%

55.6 1285.0

55.6 55.6 55.6 1285.0 55.6 1285.0 1285.0 1285.0

6% 6% 6%

587.0 1149.4

587.0 587.0 587.0 1149.4 587.0 1149.4 1149.4 1149.4

5% 5% 5% 5% 5%

1491.2

1491.2 1491.2 1491.2 1491.2

7% 7%

6%

6%

4% 4% 4%

4% 4%

T6(OH)4Na2+2+2 (2c) TT66(OH) +2 (2c) (OH)444Na Na222+2 (2c) +2 (2c) (OH) Na T66(OH) T 4Na 2 (2c) 4Na2+2 (2c) TT66(OH) (OH)Table 4Na2+2 (2c) 4. Cont. +2 T6(OH)4Na2 (2c)

Species

Structures

+ (c) T4(OH) 4Na+ (c) T4 (OH) 4 Na T4(OH) 4Na+ (c)

T4(OH)4Na+ (c) T4(OH)4Na+ (c)

T7(OH)Na2+2 (4c)

+2 +2 (4c) T7 (OH)Na TT77(OH)Na 2 22+2 (4c) (OH)Na (4c) 7(OH)Na T 7(OH)Na 2+2(4c) (4c) (4c) TT (OH)Na 7T (OH)Na(4c) (4c) T77(OH)Na

T7(OH)Na (4c)

T6(OH)6Na2+2+2 (1c)

TT66(OH) 6Na2+2 (1c) (OH) 6Na2++2 (1c) (1c) T6 (OH) 6 Na 2 2++2(1c) 6(OH) 6Na TTT66(OH) 6Na (1c) 6Na (1c) + +(1c) (OH) 6Na (1c) T6(OH) T6 (OH) Na + 6 T6(OH)6Na (1c)

T8(OH)4Na++ (3c) TT88(OH) (OH)44Na Na+ (3c) (3c)

+ (3c) T8(OH) 4Na+ (3c) T8 (OH) 4 Na

2.2. Unsaturated Unsaturated Polyester Polyester Composite Composite Materials Materials 2.2. 2.2. Unsaturated Unsaturated Polyester Polyester Composite Composite Materials Materials 2.2. Different combinations of fire retardant components components were were introduced introduced in in the the UP UP composite composite in in 2.2. Unsaturated Polyester Composite Materials Different Different combinations combinations of of fire fire retardant retardant components components were were introduced introduced in in the the UP UP composite composite in in Different combinations of fire retardant order to study the synergy effect between the aluminium trihydroxide and silsesquioxane oligomers to synergy effect the trihydroxide and silsesquioxane oligomers order to study study the theof synergy effect between between the aluminium aluminium trihydroxide and silsesquioxane oligomers Differentorder combinations fire retardant components were introduced in the UP composite in order to study the synergy effect between the aluminium trihydroxide and silsesquioxane oligomers (Table 5). 5). In In all all cases, cases, the the total total filler filler content content was was 60% 60% w/w. w/w. (Table (Table 5). In In all all effect cases, the the total filler filler content was 60% 60% w/w. order to study(Table the synergy between thecontent aluminium trihydroxide and silsesquioxane oligomers 5). cases, total was w/w. Table Compositions of the unsaturated composites and fire retardant behavior. (Table 5). In all cases, the5. total filler content was 60%polyester w/w. (UP) Table Table 5. 5. Compositions Compositions of of the the unsaturated unsaturated polyester polyester (UP) (UP) composites composites and and fire fire retardant retardant behavior. behavior. Table 5. Compositions of the unsaturated polyester (UP) composites and fire retardant behavior.

MAPTMS Limiting Oxygen Mechanical Resistance of MAPTMS Limiting Oxygen Mechanical of Composite Resin ATH POSS MAPTMS Limiting Oxygen Mechanical Resistance Resistance of Composite Resin ATH POSS (Synthesis Time) Index (LOI) Combustion Char (kg/cm MAPTMS Limiting Oxygen Mechanical Resistance of22) Table 5. Compositions of the unsaturated polyester (UP) composites and fire retardant behavior. Composite Resin ATH POSS (Synthesis Time) Index (LOI) Combustion Char Composite Resin ATH POSS (Synthesis Time) Index (LOI) Combustion--Char (kg/cm (kg/cm22)) R100 100% 0% 0% 0% 21% (Synthesis Time) Index (LOI) Combustion--Char (kg/cm ) R100 100% 0% 0% 0% 21% R100 100% 0% 0% 0% 21% --R40A60 40% 60% 0% 0% 44% --R100 100% 0% 0% 0% 21% --R40A60 40% 60% 0% 0% 44% --MAPTMS Limiting Oxygen Mechanical Resistance of R40A60 40% 60% 0% 0% 44% --R40A55M5-0 40% 55% 5%0% (0 POSS h) 0% 46% 0.7--± 0.3 R40A60 40% 60% 0% 44% Composite R40A55M5-0 Resin hATH 40% 55% 5% (0 0% Combustion 0.7 Char (kg/cm2 ) R40A55M5-0 hhh 40%(Synthesis 55% Time)5% 5%(24 (0 h) h) 0%Index (LOI)46% 46% 0.7 ±±± 0.3 0.3 R40A55M5-24 40% 55% h) 0% 48% 0.7 0.2 R40A55M5-0 h 40% 55% 5% (0 h) 0% 46% 0.7 ±± 0.2 0.3 R40A55M5-24 40% 55% 5% 0% 48% 0.7 R40A55M5-24 hhh 40% 55% 5% (24 (24 h) h) 0% 48% 0.7 ±± 0.1 0.2 R40A55M5-48 40% 55% 5% (48 h) 0% 50% 1.2 R100 100% 0% 0% 21% 50% — R40A55M5-24 40% 55% 5% (24 h) 0% 48% 0.7 0.2 R40A55M5-48 hhh0% 40% 55% 5% (48 h) 0% 1.2 ±±± 0.1 R40A55M5-48 40% 55% 5% (48 h) 0% 50% 1.2 0.1 R40A55M5-96 h60% 40% 40% 55% 5% (48 (96 h) h) 0% 50% 2.3 ±± 0.1 0.2 R40A55M5-48 h 55% 5% 0% 50% 1.2 R40A60 40% 0% 0% 44% — R40A55M5-96 40% 55% 5% (96 0% 50% 2.3 R40A55M5-96 hhh 40% 55% 5%(264 (96 h) h) 0% 50% 2.3 ±±± 0.2 0.2 R40A55M5-264 40% 55% 5% h) 0% 50% 2.8 0.2 R40A55M5-96 hh 40% 55%(0 h) 5% (96 h) 0% 50% 2.30.3 ± 0.2 R40A55M5-0 hR40A55M5-264 40% 55% 40% 5% 0% 46% 50% 0.7 ± 55% 5% 0% 2.8 R40A55M5-264 h 40% 55% 5% (264 (264 h) h) 0% 50% 2.8 ±±± 0.2 0.2 R40A55P5 40% 55% 0% 5% 51% 0.8 0.1 h 40% 55% 5% (264 h) 0% 50% 2.8 ± 0.2 R40A55M5-24 hR40A55M5-264 40% 55% 5% (24 h) 0% 48% 0.7 ± 0.2 R40A55P5 40% 55% 0% 5% 51% 0.8 R40A55P5 40% 55% 0% 5% 51% 0.8 ±± 0.1 0.1 55%(48 h) 0% 0% 5% 0.80.1 ± 0.1 R40A55M5-48 h R40A55P5 40% 55% 40% 5% 50% 51% 1.2 ±

R40A55M5-96 h 40%effectivity 55% of aluminium 5% (96 h) trihydroxide 0% as a fire retardant 50% 2.3related ± 0.2 to a high The for polymers polymers is is The effectivity of trihydroxide as for to The effectivity of aluminium aluminium trihydroxide as aaa fire fire retardant retardant for polymers polymers is is related related to aaa high high R40A55M5-264load h The 40%effectivity 55%When 5%wt (264 h) 0% as 50%the pure 2.8 ± 0.2 to of aluminium trihydroxide fire retardant for related high percentage. 60 % ATH was incorporated into UP resin, the limiting oxygen percentage. When 60 wt % ATH was incorporated into the pure UP resin, the limiting R40A55P5 load 40% 55% 0% 5% 51% 0.8 ± 0.1 load percentage. When 60 wt % ATH was incorporated into the pure UP resin, the limiting oxygen oxygen

load When 60 wt increased % ATH was incorporated into thethough pure UP the limiting oxygen indexpercentage. (LOI) was was significantly significantly from 21% to to 44%, 44%, even theresin, combustion residue had index index (LOI) (LOI) was was significantly significantly increased increased from from 21% 21% to to 44%, 44%, even even though though the the combustion combustion residue residue had had index (LOI) increased from 21% even though the combustion residue had not enough mechanical resistance to be manipulated. The effect on the fire behaviour of the partial not enough mechanical resistance to be manipulated. The effect on the fire behaviour of the partial not enough mechanical resistance to be manipulated. The effect on the fire behaviour of the partial not enough mechanical resistance be manipulated. Thedifferent effect the fire times behaviour of the substitution of 55 wt wt % % of of ATH for fortosilsesquioxane silsesquioxane with reaction and consequently consequently The effectivity of aluminium trihydroxide as a fire retardant foronpolymers is related to partial a high substitution substitution of of 55 wt wt % % of of ATH ATH for for silsesquioxane silsesquioxane with with different different reaction reaction times times and and consequently consequently substitution of ATH with different reaction times and different condensation degrees was evaluated. A further increase in the LOI value (up to 50%) was load percentage. When 60 wt % ATH was incorporated into the pure UP resin, the limiting oxygen different condensation degrees was evaluated. A further increase in the LOI value (up to 50%) different condensation condensation degrees degrees was was evaluated. evaluated. A A further further increase increase in in the the LOI LOI value value (up (up to to 50%) 50%) was was different was observed with the condensation time until 48 h of reaction time, the increase being significant as only observed with the condensation time until 48 hh44%, of reaction time, the increase being significant as only index (LOI) was significantly increased from 21% to even though the combustion residue had observed with the condensation time until 48 of reaction time, the increase being significant as only observed withhad the condensation time until 48 h of reactionSimilar time, theLOI increase being significant onlya 5% of of ATH ATH been substituted substituted by silsesquioxane. silsesquioxane. values were detectedasfor for 5% had been by Similar LOI values were detected a 5% of ATH had been substituted by silsesquioxane. Similar LOI values were detected for not enough mechanical resistance to be manipulated. The effect on the fire behaviour of the partial 5% of ATH had been substituted by silsesquioxane. Similar LOI values were detected for aa

substitution of 5 wt % of ATH for silsesquioxane with different reaction times and consequently different condensation degrees was evaluated. A further increase in the LOI value (up to 50%) was observed with the condensation time until 48 h of reaction time, the increase being significant as only 5% of ATH had been substituted by silsesquioxane. Similar LOI values were detected for a commercial completely condensed POSS (R40A55P5). Therefore, a complementary effect between aluminium trihydroxide and silsesquioxane fire retardants was observed. A different behaviour was also observed with the char strength. The composite prepared with only ATH as the fire retardant produced char that did not support their own weight (Figure 12). However, the composites prepared with a combination of ATH and silsesquioxane oligomer form char with enough consistency to measure their mechanical resistance. As showed in Table 5, the mechanical

Materials 2017, 10, 56710, 567   Materials 2017, 10, 567  Materials 2017, Materials 2017, 10, 56710, 567 Materials 2017, Materials 2017, 10, 567

11 of 14 11 of 14  11 of 14 11 of 14 11 of of 14 14 11

commercial completely condensed POSS (R40A55P5). Therefore, a complementary effect between commercial completely condensed POSS (R40A55P5). Therefore, aa complementary effect between commercial  completely  condensed  POSS  (R40A55P5).  Therefore,  complementary  effect  between  commercial completely condensed POSS (R40A55P5). Therefore, aa  complementary effect between commercial completely condensed POSSPOSS (R40A55P5). Therefore, complementary effecteffect between aluminium trihydroxide and silsesquioxane fire retardants was observed. commercial completely condensed (R40A55P5). Therefore, a complementary between aluminium trihydroxide and silsesquioxane fire retardants was observed. aluminium aluminium trihydroxide and silsesquioxane fire retardants was observed.  trihydroxide and silsesquioxane fire retardants was observed. aluminium trihydroxide and silsesquioxane fire retardants was observed. A different behaviour also observed the char strength. The composite prepared with aluminium trihydroxide andwas silsesquioxane firewith retardants was observed. A different behaviour was also with the char strength. The composite with A different behaviour was also observed with the char strength. The composite prepared with  A different behaviour was observed also observed with thedid char strength. The composite prepared with Aonly different behaviour was also observed with the char strength. The composite prepared with with ATH as the fire retardant produced char that not support their ownprepared weight (Figure 12). A different behaviour was also observed with the char strength. The composite prepared Materials 2017, 10, 567 11 of 14 only ATH as the fire retardant produced char that did not support their own weight (Figure 12). only  ATH  as  the  fire  retardant retardant  produced  char  that not did support not  support  their  own  weight  (Figure  12).  only ATH as the fire produced char that did not support their own weight (Figure 12). only only ATH as the fire retardant produced char that did their own weight (Figure 12). However, the composites prepared with a combination of ATH and silsesquioxane oligomer form ATH as the fire retardant produced char that did not support their own weight (Figure 12). However, the composites prepared with combination of ATH ATH and silsesquioxane oligomer form form However, composites prepared with aa combination of and silsesquioxane oligomer form However, the composites prepared with a combination of ATH and silsesquioxane oligomer form  However, the composites prepared with aa combination of and oligomer charthe with enough consistency to measure their mechanical resistance. As showed in Table 5, the However, the composites prepared with combination of ATH ATH and silsesquioxane silsesquioxane oligomer form char with enough consistency to measure their mechanical resistance. As showed in Table 5, the char  with  enough  consistency  to  measure measure  their  mechanical  resistance.  As cage showed  in was Table  5,  the the  char with enough consistency to their mechanical resistance. As showed in Table 5, resistance ofresistance the combustion residue of composites with open cage silsesquioxanes increased char char with enough consistency to measure their mechanical resistance. As showed in Table 5, the mechanical of the combustion residue of composites with open silsesquioxanes was with enough consistency to measure their mechanical resistance. As showed in Table 5, the mechanical resistance of the combustion residue of composites with open cage silsesquioxanes was mechanical resistance of the combustion residue of composites with open cage silsesquioxanes was mechanical resistance of the combustion residue of composites with open cage silsesquioxanes was  mechanical resistance of the combustion residue of composites with open cage silsesquioxanes was with the condensation reaction time. This behaviour was not correlated with the full polyhedral increased with the condensation reaction time.ofThis behaviour not correlated with thewas full mechanical resistance of the combustion residue composites withwas open cage silsesquioxanes increased with the condensation reaction time. This behaviour not correlated with the full increased  with  the  condensation  reaction  time.  This  behaviour  was  not  correlated  with  the  full  increased the reaction time. This behaviour was not correlated with the condensated methacryl POSS. Thus, the improvement of the was mechanical properties ofproperties the increased with with the condensation reaction time. This the behaviour was not correlated with thecombustion full polyhedral condensated methacryl POSS. Thus, improvement of the mechanical the increased with the condensation condensation reaction time. This behaviour was not correlated with theoffull full polyhedral condensated methacryl POSS. Thus, the improvement of the mechanical properties of the polyhedral condensated methacryl POSS. Thus, the improvement of the mechanical properties of the  polyhedral condensated methacryl POSS. Thus, the improvement of the mechanical properties ofopen the residue would be related to highly condensed silsesquioxane oligomers in open polyhedral structures. polyhedral condensated methacryl POSS. Thus, the improvement of the mechanical properties of the combustion residue would be related to highly condensed silsesquioxane oligomers in polyhedral condensated methacryl POSS. Thus, the improvement of the mechanical properties of the combustion residue would be related to ofhighly highly condensed silsesquioxane oligomers in open open combustion residue would be to condensed silsesquioxane oligomers in combustion  residue  would  be  related  to  highly highly  condensed  silsesquioxane  oligomers  in  open  combustion residue would be related to condensed silsesquioxane oligomers in open The presence of open-cage silsesquioxane oligomers in composites allowed between polyhedral structures. Therelated presence open-cage silsesquioxane oligomers ininteractions composites allowed combustion residue would be related to highly condensed silsesquioxane oligomers in open polyhedral structures. The presence of open-cage silsesquioxane oligomers in composites allowed polyhedral structures.  The  presence presence  of open‐cage  silsesquioxane oligomers in composites allowed  polyhedral structures. The of open-cage silsesquioxane oligomers in composites allowed components during that of improved thesilsesquioxane mechanical properties of in thecomposites char formed and avoided polyhedral structures. The firing presence open-cage oligomers allowed interactions between components during firing that improved theoligomers mechanical ofallowed the char polyhedral structures. The presence of open-cage silsesquioxane in properties composites interactions between components during firing that improved the mechanical properties of the interactions between components during firing that improved the mechanical properties of the char  interactions between components during firing that improved the mechanical properties ofchar the char dropping, as was verified by hot stage microscopy (Table 6). interactions between components during firing that improved the mechanical properties of the char formed and avoided dropping, as was verified by hot stage microscopy (Table 6). interactions between components during firing that improved the mechanical properties of the char formed and avoided avoided dropping, as was was verified by hot hot stage microscopy (Table(Table 6). 6). formed and dropping, as by microscopy (Table 6). formed and avoided dropping, as was verified by hot stage microscopy (Table 6).  formed and dropping, as was by hot microscopy formed and avoided avoided dropping, asverified was verified verified bystage hot stage stage microscopy (Table 6).

Figure Figure 12. 12. Combustion Combustion residues residues obtained obtained from from the the limiting limiting oxygen oxygen index index (LOI) (LOI) measurement. measurement. Figure 12. Combustion residues obtained from the limiting oxygen index (LOI) measurement. Figure 12. Combustion residues obtained from the limiting oxygen index (LOI) measurement.  Figure 12. Combustion residues obtained from the limiting oxygen index (LOI) measurement. FigureFigure 12. Combustion residues obtained from the limiting oxygen index (LOI) measurement. Figure 12. 12. Combustion Combustion residues residues obtained obtained from from the the limiting limiting oxygen oxygen index index (LOI) (LOI) measurement. measurement. Table measurements. Table 6. 6. Silhouettes Silhouettes of of the the samples samples during during the the hot hot stage stage microscopy microscopy measurements. TableTable 6. Silhouettes Silhouettes of the theof samples duringduring the hot hot stage microscopy measurements. Table Table 6. Silhouettes of the samples during the hot stage microscopy measurements.  6. of samples during the stage microscopy measurements. 6. Silhouettes the samples the hot stage microscopy measurements. Table 6. Silhouettes of the samples during the hot stage microscopy measurements.

Temperature (°C) Temperature (◦ C) Temperature (°C) Temperature (°C)  (°C) Temperature Temperature Temperature (°C) (°C)

R100

R100 R100 R100 R100

R100

 

R+M R+ +M R ++ M MR R R+M M

25 25 25 25  25 25 25   350 350 350 350 350  350 350   380 380 380  380 380 380 380

  400 400 400 400 400  400 400   800 800 800 800 800 800 800 

----—

------‐‐‐ 

    The thermal stability of the UP composites was also evaluated by differential thermal analysis The thermal stability of the composites was also evaluated by differential thermal analysis The(Figure thermal stability ofUP the UP composites was also evaluated byreactions differential thermal analysis The thermal stability of the UP composites was also evaluated by differential thermal analysis (DTA) 13). Although highly exothermic complex combustion of the pure analysis polyester The thermal stability ofofthe UP composites was also evaluated bybydifferential thermal The thermal stability the UP composites was also evaluated differential thermal analysis (DTA) (Figure 13). Although highly exothermic complex combustion reactions of the pure polyester (DTA) (Figure 13). Although highly exothermic complex combustion reactions ofpure the pure polyester (DTA) (Figure 13). Although highly exothermic complex combustion reactions of the polyester (DTA) (Figure 13). Although highly exothermic complex combustion reactions of the pure polyester (DTA) (Figure 13). Although highly exothermic complex combustion reactions of the pure polyester polymer (R100) were registered between 300 ◦ C and 400 ◦ C, a noteworthy decrease of the exothermic character was observed by the introduction of aluminium trihydroxide (R40A60). The improvement in

Materials 2017, 10, 567

12 of 14

Materials 2017, 10, 567

12 of 14

polymer (R100) were registered between 300 °C and 400 °C, a noteworthy decrease of the exothermic character was observed by the introduction of aluminium trihydroxide (R40A60). The improvement the thermal behaviour of the composites was slightly increased the combination silsesquioxanes in the thermal behaviour of the composites was slightlybyincreased by theofcombination of and aluminium trihydroxide as fire retardants (R40A55M5-264h). silsesquioxanes and aluminium trihydroxide as fire retardants (R40A55M5-264h).

Figure 13. 13. Differential thermal analyses analyses of of three three of of the the studied studied samples. samples. Figure Differential thermal

3. Materials Materials and and Methods Methods 3. 3.1. Materials 3-Methacryloxypropyltrimethoxysilane(MAPTMS) (MAPTMS) (Sigma-Aldrich Química, S.A, Madrid, 3-Methacryloxypropyltrimethoxysilane (Sigma-Aldrich Química, S.A, Madrid, Spain), Spain), was selected as the precursor the silsesquioxane oligomer, it has reactive unsaturations was selected as the precursor of theof silsesquioxane oligomer, as it as has reactive unsaturations in Inc. ( Hattiesburg, MS, in the methacrylate group. Methacryl POSS was supplied by Hybrid Plastics the methacrylate group. POSS was supplied by Hybrid Plastics Inc. (Hattiesburg, USA). Aluminium trihydroxide (ATH) was kindly Materials MS, USA). Aluminium trihydroxide (ATH) was kindlysupplied suppliedby by Huber Huber Engineered Engineered Materials (Ostende, Belgium). Belgium). (Ostende, Unsaturatedpolyester polyesterresin resin kindly supplied by Spain, Ferro S.A. Spain, S.A. (Castellon, Spain), Unsaturated waswas kindly supplied by Ferro (Castellon, Spain), (obtained (obtained by the reaction between maleic phthalic anhydride, phthalic and anhydride, ethylene glycol, and by the reaction between maleic anhydride, anhydride, ethyleneand glycol, and later diluted later dilutedAccelerator in styrene).NL-51P, Accelerator NL-51P, a cobalt(II) 2-ethylhexanoate wt Co,mixture, in a solvent in styrene). a cobalt(II) 2-ethylhexanoate 6% wt Co, in 6% a solvent was mixture, was employed as anand accelerator, a methyl ketone peroxide (MEKP) employed as an accelerator, Butanox and LA, Butanox a methylLA, ethyl ketone ethyl peroxide (MEKP) solution, as solution, as Both an initiator. supplied Akzo Nobel Chemicals S.A. (El Prat de an initiator. additivesBoth wereadditives supplied were by Akzo Nobel by Chemicals S.A. (El Prat de Llobregat, Spain). Llobregat, Spain). 3.2. Preparation of Silsesquioxane Oligomers 3.2. Preparation of Silsesquioxane Oligomers The synthesis was performed under acid-catalysed conditions since, according to the literature [6], condensation yieldswas smaller-sized than those obtained under basic conditions. Inliterature a typical The synthesis performedmolecules under acid-catalysed conditions since, according to the synthesis, 20 g (0.08 moles) of MAPTMS was poured a 200 mL vessel. silane was stirred [6], condensation yields smaller-sized molecules than into those obtained underThe basic conditions. In a while g (0.36 moles) of a 0.01 N HCl aqueous solution was slowly theThe synthesis typical6.5synthesis, 20 g (0.08 moles) of MAPTMS was poured into a added. 200 mLThen, vessel. silane was performed with mole ratio 4.5:1Nbecause this was the ratio thatslowly led to added. more condensed stirred while 6.5a gH(0.36 of a=0.01 HCl aqueous solution was Then, the 2 O:Si moles) structures. Theperformed mixture was constantly Materials extracted at reaction times 24toh,more 48 h, synthesis was with a H2O:Sistirred. mole ratio = 4.5:1were because this was the ratio thatof led 72 h, 96 h, 168 h, and 264 h. mixture The solvent evaporated off under vacuum a clearatviscous product condensed structures. The waswas constantly stirred. Materials wereand extracted reaction times was of 24obtained. h, 48 h, 72 h, 96 h, 168 h, and 264 h. The solvent was evaporated off under vacuum and a clear viscous product was obtained. 3.3. Preparation of Polymer Composites

3.3. Preparation of Polymer Composites The unsaturated polyester resin was previously activated with 1% of Accelerator NL-51P. The polymer composites were resin prepared by mixing activated 40 wt % with of activated unsaturated polyester The unsaturated polyester was previously 1% of Accelerator NL-51P. The resin with 60 wt % of a fire-retardant mixture. Different combinations of fire retardants were tested. polymer composites were prepared by mixing 40 wt % of activated unsaturated polyester resin with The polymerization reaction was initiated by adding 2% of Butanox then thewere mixture was cast 60 wt % of a fire-retardant mixture. Different combinations of LA fire and retardants tested. The into 20 mm × 80 mm × 6 mm moulds.

Materials 2017, 10, 567

13 of 14

3.4. Silsesquioxane Oligomer Structural Characterisation The MAPTMS hydrolysis and condensation reactions were monitored by nuclear magnetic resonance (NMR) spectroscopy of the 1 H, 13 C, and 29 Si nuclei. This technique allowed the different degrees of silsesquioxane condensation during the reaction to be quantified. A Varian VNMR System 500 MHz spectrometer (Varian Inc , Palo Alto, CA, USA) was used. The chemical shifts were expressed in parts per million (δ ppm) with respect to tetramethylsilane (TMS). Deuterated chloroform (CDCl3 ) was used as an internal reference. The study of the chemical species present in the synthesised oligomer was completed by Fourier transform infrared spectroscopy (FTIR) and mass spectrometry. The infrared spectra were obtained on a THERMO-NICOLET 6700 instrument (Thermo Fischer Scientific Inc., Waltham, Massachusetts), between 400 and 4000 cm−1 , recording 16 scans at 4 cm−1 resolution. The electrospray ionization time-of-flight mass spectra (ESI-TOF-MS) were obtained on a QTOF I (quadrupole–hexapole–TOF) hybrid spectrometer (Micromass UK Ltd., Wilmslow, UK) with a Z-spray electrospray orthogonal interface. 3.5. Polymer Composite Characterisation DTA was used in order to study the thermal resistance of the polymer and composite materials. The DTA measurements were performed with a TGA/SDTA 851e thermal analysis apparatus (Mettler Toledo International Inc., L'Hospitalet de Llobregat, Spain) between 30 ◦ C and 800 ◦ C with a heating rate of 5 ◦ C/min under air. Around 15 mg of samples were introduced in platinum crucibles, with ignited alumina (Al2 O3 ) chosen as the reference material. The evolution of the shape of the specimens during firing was evaluated by hot stage microscopy. The experiments were run on a MISURA 2 (Expert System Solutions, Modena, Italy) with a heating rate of 25 ◦ C/min. Images of the specimen silhouettes were taken every 10 ◦ C. The LOI was obtained following the ISO 4589:2001 procedure, which involves measuring the minimum oxygen concentration required for self-sustained combustion of the samples [20]. The mechanical strength of the combustion residue was determined from prism-shaped test 6 mm × 20 mm × 80 mm pieces obtained after firing in LOI testing. The bending strength was determined by the three-point bending method with an universal testing machine 6027 (Instron, Norwood, MA, USA). 4. Conclusions Silsesquioxane oligomers were synthesised in open structures by the hydrolysis and condensation reactions of MAPTMS. Different characterisation techniques were used to study the arising molecular structures. The presence of hydrogen bonds between the carbonyl groups and the hydroxyls was identified by both NMR and FT-IR spectroscopy and could be hindering the closed-cage structures. A complementary effect was observed between aluminium trihydroxide and silsesquioxanes in the thermal and fire retardant properties of the UP composites. Firstly, the exothermicity of the combustion reaction was decreased. Secondly, the minimum oxygen concentration required to sustain combustion (LOI) was significantly increased. Furthermore, obtaining open silsesquioxane structures instead of closed cage structures also improved the mechanical properties of the char obtained after firing. Author Contributions: V.S. and Y.B. conceived and designed the experiments; A.G. and S.M. performed the experiments; S.M. and V.S. analyzed the data; Y.B. and A.G. wrote the paper. Conflicts of Interest: The authors declare no conflict of interest.

References 1. 2.

Laoutid, F.; Bonnaud, L.; Alexandre, M.; López-Cuesta, J.-M.; Dubois, P.H. New prospects in flame retardant polymer materials: From fundamentals to nanocomposites. Mater. Sci. Eng. 2009, 63, 100–125. [CrossRef] Wilkie, C.A.; Morgan, A.B. (Eds.) Fire Retardancy of Polymeric Materials, 2nd ed.; CRC Press: Boca Raton, FL, USA, 2010.

Materials 2017, 10, 567

3. 4. 5.

6. 7. 8. 9.

10.

11. 12. 13. 14. 15. 16.

17. 18. 19.

20.

14 of 14

Zhou, W.; Yang, H.; Guo, X.; Lu, J. Thermal degradation behaviors of some branched and linear polysiloxanes. Polym. Degrad. Stab. 2006, 91, 1471–1475. [CrossRef] Zhanga, W.; Camino, G.; Yang, R. Polymer/polyhedral oligomeric silsesquioxane (POSS) nanocomposites: An overview of fire retardance. Prog. Polym. Sci. 2017, 67, 77–125. [CrossRef] Shea, K.J.; Moreau, J.; Loy, D.A.; Corriu, R.J.; Boury, P.B. Bridged Polysilsesquioxanes. Molecular-Engineering Nanostructured Hybrid Organic-Inorganic Materials. In Functional Hybrid Materials; Gomez Romero, P., Sanchez, C., Eds.; Wiley-VCH: Weinheim, Germany, 2004; ISBN 978-3-527-30484-4. Brinker, C.J.; Scherer, G.W. Sol–Gel Science, the Physics and Chemistry of Sol–gel Processing; Academia Press: Boston, MA, USA, 1990. Li, G.; Wang, L.; Ni, H.; Pittman, C.U., Jr. Polyhedral Oligomeric Silsesquioxane (POSS). Polymers and Copolymers: A Review. J. Inorg. Organomet. Polym. 2001, 11, 123–154. [CrossRef] Engelhardt, G.; Michel, D. High Resolution Solid-State NMR of Silicates and Zeolites; John Wiley & Sons: New York, NY, USA, 1987. Pantoja, M.; Diaz-Benito, B.; Velasco, F.; Abenojar, J.; Real, J.C.D. Analysis of hydrolysis process of γ-methacryloxypropyltrimethoxysilane and its influence on the formation of silane coatings on 6063 aluminum alloy. Appl. Surf. Sci. 2009, 255, 6386–6390. [CrossRef] Soppera, O.; Croutxé-Barghorn, C. Real-time Fourier transform infrared study of free-radical UV-induced polymerization of hybrid sol–gel. I. Effect of silicate backbone on photopolymerization kinetics. J. Polym. Sci. A Polym. Chem. 2003, 41, 716–724. [CrossRef] Chan, C.-K.; Chu, I.M. Effect of hydrogen bonding on the glass transition behaviour of poly(acrylic acid)/silica hybrid materials prepared by sol–gel process. Polymer 2001, 42, 6089–6093. [CrossRef] Landry, C.J.T.; Coltrain, B.K.; Wesson, J.A.; Lippert, J.L.; Zumbulyadis, N. In situ polymerization of tetraethoxysilane in polymers: Chemical nature of the interactions. Polymer 1992, 33, 1496–1506. [CrossRef] Sugahara, Y.; Okada, S.; Sato, S.; Kuroda, K.; Kato, C. 29 Si-NMR study of hydrolysis and initial polycondensation processes of organoalkoxysilanes. II. Methyltriethoxysilane. J. Non Cryst. Solids 1994, 167, 21–28. [CrossRef] Babonneau, F.; Maquet, J. Nuclear magnetic resonance techniques for the structural characterization of siloxane–oxide hybrid materials. Polyhedron 2000, 19, 315–322. [CrossRef] Brunet, F. Polymerization reactions in methyltriethoxysilane studied through 29 Si NMR with polarization transfer. J. Non-Cryst. Solids 1998, 231, 58–77. [CrossRef] Hanjiang, D.; Manho, L.; Ruthanne, D.T.; Zhengping, Z.; Reidy, R.F.; Mueller, D.W. Methyltrimethoxysilane sol–gel polymerization in acidic ethanol solutions studied by 29 Si NMR spectroscopy. J. Sol Gel Sci. Technol. 2003, 28, 5–14. [CrossRef] Brunet, F.; Cabane, J. Populations of oligomers in sol–gel condensation. J. Non Cryst. Solids 1993, 163, 211–225. [CrossRef] Devereux, F.; Boilot, J.P.; Chaput, F. Sol–gel condensation of rapidly hydrolyzed silicon alkoxides: A joint 29 Si NMR and small-angle x-ray scattering study. Phys. Rev. A 1990, 41, 6901–6909. [CrossRef] Eisenberg, P.; Erra-Balsells, R.; Ishikawa, Y.; Lucas, J.C.; Nonami, H.; Williams, R.J.J. Silsesquioxanes derived from the bulk polycondensation of [3-(Methacryloxy)propyl]trimethoxysilane with concentrated formic acid: Evolution of molar mass distributions and fraction of intramolecular cycles. Macromolecules 2002, 35, 1160–1174. [CrossRef] UNE-EN ISO 4589-2. Plastics—Determination of Burning Behaviour by Oxygen Index—Part 2: Ambient-Temperature Test; AENOR: Madrid, Spain, 2001. © 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/).