Thermal Degradation Mechanism of a Thermostable Polyester Stabilized with an Open-Cage Oligomeric Silsesquioxane

materials

Article Thermal Degradation Mechanism of a Thermostable Polyester Stabilized with an Open-Cage Oligomeric Silsesquioxane

Yolanda Bautista * ID , Ana Gozalbo, Sergio Mestre and Vicente Sanz

University Institute of Ceramic Technology, Jaume I University, 12006 Castellón, Spain; [email protected] (A.G.); [email protected] (S.M.); [email protected] (V.S.) * Correspondence: [email protected]; Tel.: +34-608-066-579

Received: 24 October 2017; Accepted: 20 December 2017; Published: 24 December 2017

Abstract: A polyester composite was prepared through the polymerization of an unsaturated ester resin with styrene and an open-cage oligomeric silsesquioxane with methacrylate groups. The effect of the open-cage oligomeric silsesquioxane on the thermal stability of the thermostable polyester was studied using both thermogravimetric analysis and differential thermal analysis. The results showed that the methacryl oligomeric silsesquioxane improved the thermal stability of the polyester. The decomposition mechanism of the polyester/oligomer silsesquioxane composite was proposed by Fourier transform infrared spectroscopy (FTIR) analysis of the volatiles.

Keywords: oligomeric silsesquioxane; thermal degradation; open cage structure; thermostable polyester; FTIR volatiles analysis

1. Introduction Unsaturated polyester resins are the most frequently used organic matrices in composite materials. Therefore, their use has extended to numerous fields such as the maritime, automotive, and aeronautical transport sectors, the energy field (in the wind turbine industry), or the building sector, and in industries manufacturing panels, bathroom components, or electrical wiring [1]. Unfortunately, the polyesters prepared with these resins are flammable materials with spontaneous flame propagation in the presence of oxygen. The high cost in human lives and material damage that can result from a fire has greatly promoted the development of flame retardants. These products usually act by suppressing or delaying the physical-chemical processes that occur during fire by eliminating some of the components of the fire cycle [2,3]. The nature of the flame retardants used commercially in polyesters varies greatly. Originally, halogenated compounds were used, but they were soon removed due to the toxicity of the gases generated in their decomposition. Subsequently, aluminum trihydroxide (ATH) was introduced, thus avoiding the toxicity of the gases, but with the disadvantage of a very low consistency of the polymer ashes after fire. Recently, considerable attention has been paid to the effect of silsesquioxane, usually in cage structures, on a polymer’s thermal resistance [4]. An improvement has been found in epoxy [5,6], polycarbonate [7,8], polystyrene [9], polyurethane [10], and polymethyl methacrylate [11] polymers. In previous work [12,13], our research group improved both the thermal and fire behavior of a polyester by adding silsesquioxane oligomers together with ATH. The LOI (limiting oxygen index) value was increased from 21% for the original polyester to 51% for the samples to which ATH and silsesquioxane were simultaneously added. It was also observed that the incorporation of an open-cage silsesquioxane to the mixture of polyester and ATH produced a fourfold increase in the mechanical

Materials 2018, 11, 22; doi:10.3390/ma11010022 www.mdpi.com/journal/materials Materials 2017, 11, 22 2 of 13 cageMaterials silsesquioxane2018, 11, 22 to the mixture of polyester and ATH produced a fourfold increase in2 of the 13 mechanical resistance of the ashes. In contrast, there was no increase in mechanical resistance when aresistance close-cage of form the ashes. of the Insilses contrast,quioxane there was was added no increase instead. in mechanical resistance when a close-cage formIn of this the work, silsesquioxane the effect wasof the added open instead.-cage oligomeric silsesquioxane on the thermal stability of the polyesterIn this was work, studied the effect by using of the both open-cage thermogravimetric oligomeric silsesquioxaneand differential on thermal the thermal analysis, stability with of a simultaneousthe polyester wasanalysis studied of the by vol usingatiles both by thermogravimetricFourier transform infrared and differential spectroscopy thermal (FTIR). analysis, A thermal with a decompositionsimultaneous analysis mechanism of the was volatiles proposed by Fourier for the transform polyester/oligomer infrared spectroscopy open-cage (FTIR). silsesquioxane A thermal composite.decomposition Although mechanism the thermal was degradation proposed for of polyesters the polyester/oligomer has been extensively open-cage studied silsesquioxane over the last thirtycomposite. years Although[14–25], the the effect thermal of open degradation-cage silsesquioxane of polyesters hashas beenyet to extensively be assessed. studied Even overthough the the last analyzedthirty years polyesters [14–25], were the effect quite of different, open-cage the silsesquioxanesame three general has yet processes to be assessed. were found: Even degradation though the ofanalyzed the main polyesters chain of the were polyester, quite different, degradation the same of the three styrene general interconnections, processes were and found: oxidation degradation of the productsof the main generated chain of in the both polyester, of those degradationprevious ones. of theHowever, styrene the interconnections, temperature at which and oxidation every process of the proceeded,products generated as well inas both the type of those of products previous generatedones. However,, were the signifi temperaturecantly different, at which depending every process on severalproceeded, factors as well such as as the the type type of productsof glycol, generated, the nature were of signiﬁcantlythe aromatic different, or the saturated depending diacid on severals, the initiator,factors such or the as curing the type process of glycol, [15]. the nature of the aromatic or the saturated diacids, the initiator, or the curing process [15]. 2. Results and Discussion 2. ResultsA commercial and Discussion unsaturated polyester resin was obtained by the reaction between maleic anhydride,A commercial phthalic unsaturated anhydride, polyester and ethylene resin was glycol. obtained Then, by the the main reaction chain between of the maleic resin anhydride, had ester groups,phthalic aromatic anhydride, rings and, and ethylene unsaturated glycol. Then, sites, theas shown main chain in Figure of the 1. resin Usually, had ester polyester groups, resins aromatic are mixedrings, andwith unsaturated styrene, which sites, acts as as shown both a indiluent Figure and1. Usually, as a cross polyester-linking agent. resins Polymerization are mixed with involves styrene, thewhich reaction acts as of both the aunsaturated diluent and sites as a cross-linkingin the resin with agent. the Polymerization vinyl group of involves the styrene, the reaction as shown of thein Figureunsaturated 1. sites in the resin with the vinyl group of the styrene, as shown in Figure1.

FigureFigure 1. 1. SchemeScheme of of the the synthesis synthesis and and polymerization polymerization of of an an unsaturated unsaturated polyester polyester resin. resin.

The organic–inorganic hybrid oligomer was synthesized by the sol–gel method from 3- The organic–inorganic hybrid oligomer was synthesized by the sol–gel method from methacryloxypropyltrimethoxysilane (MAPTMS) with hydrolyzable methoxy groups, which 3-methacryloxypropyltrimethoxysilane (MAPTMS) with hydrolyzable methoxy groups, which allowed allowed the formation of siloxane bonds. The oligomer consisted mainly of mixtures of four to eight the formation of siloxane bonds. The oligomer consisted mainly of mixtures of four to eight silsesquioxane molecules containing unreacted hydroxyl groups (hydrolyzed silanes, but not yet silsesquioxane molecules containing unreacted hydroxyl groups (hydrolyzed silanes, but not yet condensated) [12]. The participation of these groups in hydrogen bonds with the methacrylate groups condensated) [12]. The participation of these groups in hydrogen bonds with the methacrylate groups prevented the condensation reactions from concluding. The oligomer exhibited a high number of prevented the condensation reactions from concluding. The oligomer exhibited a high number cycles per molecule without forming the closed polyhedral structures characteristic of POSS of cycles per molecule without forming the closed polyhedral structures characteristic of POSS (polyhedral oligomeric silsesquioxane). Figure 2 details an example of one of the molecules that (polyhedral oligomeric silsesquioxane). Figure2 details an example of one of the molecules that formed the oligomer. formed the oligomer.

MaterialsMaterials 20172018, 11, 11, 22, 22 33 of of 13 13 Materials 2017, 11, 22 3 of 13

Figure 2. Scheme of one of the molecules of the organic-inorganic hybrid oligomer [12]. FigureFigure 2. 2. SchemeScheme of of one one of of the the molecu moleculesles of of the the organic organic-inorganic-inorganic hybrid oligomer [ 12]. 2.1. Thermal Degradation of the Thermostable Polyester 2.1.2.1. Thermal Thermal Degradation Degradation of of the the Thermostable Thermostable Polyester Polyester The thermal degradation of thermostable polyesters is a complex, heterogeneous process of severalTheThe exo/endothermic thermal degradation reactions of of producing thermostable volatile po polyesterslyesters compounds. is is a complex, The weight heterogeneous loss of the sample process process was of of registeredseveralseveral exo/endothermic by thermogravimetric reactions reactions analysis producing producing, and volatile volatilethe volatiles compounds. compounds. were identified The The weight weight by infrared loss loss of of the the spectroscopy. sample sample was was Differentialregisteredregistered by bythermal thermogravimetric thermogravimetric analysis (DTA) analysis analysis, and differential, and and the the volatiles volatilesscanning were were calorimetry identified identiﬁed (DSC) by by infrared infrared were used spectroscopy. spectroscopy. to follow theDifferentialDifferential heat generated. thermal analysis analysis (DTA) (DTA) and differential scanning calorimetry calorimetry (DSC) (DSC) were were used used to to follow follow thethe heat heatThermogravimetric generated. generated. analysis can provide information not only on the thermal stability, but also on theThermogravimetricThermogravimetric reactions taking place. analysisanalysis Figure cancan provideprovide3 shows information information the thermogravimetric not not only only on on the analysisthe thermal thermal of stability, thestability, thermostable but but also also on polyesteronthe the reactions reactions in (a)taking air,taking place. and place. Figure(b) nitrogen Figure3 shows 3 atmospheres.shows the thermogravimetric the thermogravimetric In air (Figure analysis 3a), analysis of the two thermostable groupsof the thermostable of chemical polyester transformationspolyesterin (a) air, and in (a) (b) with air, nitrogen associated and (b) atmospheres. nitrogen mass loss atmospheres. In could air (Figure be appreciated. In3a), air two (Figure groups The first 3a), of group chemical two groupstook transformations place of between chemical 200transformationswith °C associated and 450 ° massCwith and lossassociated accounted couldbe mass for appreciated. 90% loss ofcould the The totalbe ﬁrstappreciated. mass group loss. took The place asymmetricfirst betweengroup took shape 200 place◦C of and the between 450peak◦C suggested200and ° accountedC and that 450 it for° Cwas 90%and the ofaccounted result the total of for massseveral 90% loss. steps.of Thethe Thetotal asymmetric second mass loss. group shape The ofwas asymmetric the registered peak suggested shape between of that the 450 itpeak was°C andsuggestedthe result570 °C, of that and several it accounted was steps. the Theresult forsecond the of remainingseveral group steps. was mass registered The loss. second As between this group last 450 peakwas◦C registered was and 570only◦ C,observedbetween and accounted 450when °C thermalandfor the570 remainingdegradation °C, and accounted mass occurred loss. for As in the thisthe remainingpresence last peak of was mass oxygen, only loss. observed it As could this correspond whenlast peak thermal was to a degradation onlycombustion observed occurredprocess when [thermal25in]. the presence degradation of oxygen, occurred it could in the correspond presence of to oxygen, a combustion it could process correspond [25]. to a combustion process [25].

(a) (b) (a) (b) FigureFigure 3. 3. ThermogravimetricThermogravimetric analysis analysis of of the the thermostable thermostable polyester polyester (a (a) )in in air, air; and and (b (b) )in in nitrogen. nitrogen. Figure 3. Thermogravimetric analysis of the thermostable polyester (a) in air, and (b) in nitrogen. The differential thermal analysis (Figure 4) showed that the group of chemical transformations occurringTheThe differential between 200 thermal °C and analysis 450 °C in (Figure (Figure the absence 44)) showedshowed of oxygen thatthat thethe were groupgroup slightly ofof chemicalchemical endothermic transformationstransformations processes, occurring between 200 ◦C and 450 ◦C in the absence of oxygen were slightly endothermic processes, whichoccurring would between likely 200 correspond °C and 450 to °C pyrolytic in the absence reactions of oxygen [14]. Whenwere slightly the thermal endothermic degradation processes, was studiedwhich would would in thelikely airlikely atmosphere, correspond correspond to both pyrolyticto pyrolyticgroups reactions of reactions chemical [14]. [ Whentransformations14]. When the thermal the thermal were degradation highly degradation exothermic, was studied was studied in the air atmosphere, both groups of chemical transformations were highly exothermic,

Materials 2018, 11, 22 4 of 13

Materials 2017, 11, 22 4 of 13 inMaterials the air 2017 atmosphere,, 11, 22 both groups of chemical transformations were highly exothermic, raising4 of 13 ◦ theraising temperature the temperature as much as much as 23 asC 23 for °C a for polymer a polymer sample sample of 15of 15 mg. mg. Although Although pyrolytic pyrolytic reactionsreactions alsoalsoraising occuroccur the inintemperature anan airair atmosphere,atmosphere, as much as theythey 23 are°areC for maskedmasked a polymer byby thethe sample highlyhighly of exothermic exothermic15 mg. Although combustioncombustion pyrolytic reactionsreactions reactions ofof ◦ thethealso volatilesvolatiles occur in [[an2626]. ].air TheThe atmosphere, higher-temperaturehigher-temperature they are masked chemicalchemical by transformationstransformationsthe highly exothermic (450–570(450–570 combustion °C)C) werewere reactions exothermicexothermic of combustioncombustionthe volatiles processes.processes. [26]. The higher-temperature chemical transformations (450–570 °C) were exothermic combustion processes.

FigureFigure 4. 4. Differential thermal analysis (DTA) of the thermostable thermostable polyester in air and and nitrogen nitrogen Figure 4. Differential thermal analysis (DTA) of the thermostable polyester in air and nitrogen atmospheresatmospheres (heating(heating raterate ofof 1515 ◦°C/min).C/min). atmospheres (heating rate of 15 °C/min). Figure 5 compares the infrared spectra of the polyester calcined at two different temperatures Figure5 compares the infrared spectra of the polyester calcined at two different temperatures with Fthatigure of 5the compares original thepolyester. infrared A spectra decrease of ofthe the polyester signals relatedcalcined to at the two ester different groups temperatures (1720, 1320, with that of the original polyester. A decrease of the signals related to the ester groups (1720, 1320, andwith 940 that cm of− 1the) and original the aliphatic polyester. groups A decrease (2930, 2860, of the and signals 1450 cm related−1) was to observed, the ester whereasgroups (1720, the signals 1320, −−1 1 −1 −1 andand 940 cm cm ) and) and the the aliphatic aliphatic groups groups (2930, (2930, 2860, 2860, and and1450 1450 cm−1 ) cm was )observed, was observed, whereas whereas the signals the of the aromatic groups (3060, 3030, 1600, 760, and 700 cm −) increased. Therefore, the thermal signals of the aromatic groups (3060, 3030, 1600, 760, and 700 cm−1 1) increased. Therefore, the thermal degradationof the aromatic of the groups polyester (3060, generated 3030, 1600, ashes 760, was and enriched 700 cm in aromatic) increased. groups Therefore, by the formation the thermal of degradation of the polyester generated ashes was enriched in aromatic groups by the formation of polyphenolicdegradation of struct the urespolyester [14]. generated ashes was enriched in aromatic groups by the formation of polyphenolicpolyphenolic structuresstructures [[1414].].

Figure 5. Fourier transform infrared (FTIR) spectra of the original thermostable polyester and the FigureashesFigure generated 5.5.Fourier Fourier after transform transform calcination infrared infrared at 390 (FTIR) (FTIR) °C spectraand spectra 450 of °C theof. originalthe original thermostable thermostable polyester polyester and the and ashes the generatedashes generated after calcination after calcination at 390 ◦atC 390 and °C 450 and◦C. 450 °C . To study the thermal degradation of the thermostable polyester, FTIR spectra of the sample To study the thermal degradation of the thermostable polyester, FTIR spectra of the sample volatilesTo study were thetaken thermal at each degradation 5 °C for calcination of the thermostable temperatures polyester, ranging from FTIR 300 spectra °C to 500 of the °C sample(Figure volatiles were taken at each 5 °C for calcination temperatures ranging from 300 °C to 500 °C (Figure volatiles6). Table were 1 enumerates taken at each the 5 species◦C for calcination corresponding temperatures to each rangingof the main from infrared 300 ◦C to peaks 500 ◦ C found. (Figure The6). 6). Table 1 enumerates the species corresponding to each of the main infrared peaks found. The Tableevolution1 enumerates of each infrared the species signal corresponding with temperature to each (giving of the maina semi infrared-quantitative peaks information found. The evolution about the evolution of each infrared signal with temperature (giving a semi-quantitative information about the ofvolatile each infraredspecies) signalis depicted with in temperature Figure 7, for (giving calcination a semi-quantitative in air and nitrogen information atmospheres. about the volatile volatile species) is depicted in Figure 7, for calcination in air and nitrogen atmospheres. species) is depicted in Figure7, for calcination in air and nitrogen atmospheres.

Materials 2017, 11, 22 5 of 13

MaterialsMaterials2018 2017, 11, 11, 22, 22 5 of5 13 of 13

FigureFigure 6. 6.FTIR FTIR spectra spectra of of the the volatile volatile species species produced produced during during the thermal the thermal degradation degradation of the of polyester the aspolyesterFigure a function 6as. FTIRa of function temperature, spectra of temperature, of thein air volatile atmosphere. in air species atmosphere. produced during the thermal degradation of the polyester as a function of temperature, in air atmosphere. TableTable 1.1. InfraredInfrared signals of of the the volatile volatile species species identified identified in inthe the thermal thermal degradation degradation of the of thepolyester. polyester. Table 1. Infrared signalsWavenumber of the volatile (cm species−1) Identifiedidentified in Volatile the thermal Sp eciesdegradation of the polyester. Wavenumber (cm−1) Identified Volatile Species 3720,Wavenumber 3620, 2358, (cm 670− 1) IdentifiedCO Volatile2 Species 3720, 3620, 2358, 670 CO2 3720,2173, 3620, 2080 2358, 670 COCO 2 2173, 2080 CO 3650, 1650, 1490 H2O 3650,2173, 1650, 14902080 CO H O 1770, 1290, 1130 Phthalates 2 1770,3650, 1290, 1650, 1130 1490 H Phthalates2O 3090,3090,1770, 2810, 2810, 1290, 2730, 2730, 17311130 1731 BenzaldehydePhthalates Benzaldehyde 3070,3070,3090, 2710, 2710, 2810, 1745, 1745, 2730, 1030 1030 1731 PhenylacetaldehydeBenzaldehyde Phenylacetaldehyde 3070,1803,1803, 2710, 1250, 1250, 1745, 890 890 1030 PhthalicPhenylacetaldehyde Phthalic anhydride anhydride 1803, 1250, 890 Phthalic anhydride As can be seen in Figure 7, the thermal degradation of the polyester in both atmospheres As can be seen in Figure7, the thermal degradation of the polyester in both atmospheres produced produced two groups of volatile species between 300 °C and 450 °C. The first group appeared around two groupsAs can of volatile be seen species in Figure between 7, the 300 thermal◦C and degradation 450 ◦C. The of first the group polyester appeared in both around atmospheres 365 ◦C and 365 °C and corresponded mainly to species such as carbon dioxide, aldehydes, and phthalates, while correspondedproduced two mainly groups to of species volatile such species as carbonbetween dioxide, 300 °C and aldehydes, 450 °C. The and first phthalates, group appeared while the around second the second one, appearing between 400 °C and 450 °C, essentially consisted of phthalic anhydride. one,365 appearing °C and corresponded between 400 mainly◦C and to 450 species◦C, essentially such as carbon consisted dioxide, of phthalic aldehydes, anhydride. and phthalates Carbon, while dioxide Carbonthe second dioxide one, was appearing registered between in both 400 the °C air and and 450 nitrogen °C, essentially atmospheres; consisted thus, of it phthaliccan be considered anhydride. was registered in both the air and nitrogen atmospheres; thus, it can be considered as a product of asCarbon a product dioxide of both was pyrolysis registered and in bothcom bustionthe air and reactions. nitrogen Carbon atmospheres; monoxide thus, and it canwater be wereconsidered also both pyrolysis and combustion reactions. Carbon monoxide and water were also detected as reaction detectedas a product as reaction of both products pyrolysis in both and atmospheres combustion, thoughreactions. their Carbon proportion monoxide was practically and water negligible were also products in both atmospheres, though their proportion was practically negligible when the assay was whendetected the assay as reaction was performed products in in both nitrogen. atmospheres , though their proportion was practically negligible performed in nitrogen. when the assay was performed in nitrogen.

(a) (b)

Figure 7. Evolution with(a )temperature of the infrared signal of each identified volatile(b) species during theFigure thermal 7. Evolution degradation with of temperaturethe polyester of (a the) in infraredair atmosphere signal of, and each (b identified) in nitrogen. volatile species during Figure 7. Evolution with temperature of the infrared signal of each identiﬁed volatile species during the thermal degradation of the polyester (a) in air atmosphere, and (b) in nitrogen. 2.2. Thermalthe thermal Degradation degradation of the of Polyester/Oligomer the polyester (a) in Silsesquioxane air atmosphere; Composite and (b) in nitrogen. 2.2.A Thermals shown Degradation in a previous of the work Polyester/Oligomer [12], the silsesquioxane Silsesquioxane oligomer Composite synthetized was a mixture of 2.2. Thermal Degradation of the Polyester/Oligomer Silsesquioxane Composite compoundsAs shown obtained in a byprevious condensation work [ of12 ],between the silsesquioxane four to eight oligomer silane molecul synthetizedes, with was a high a mixture number of compoundsAs shown obtained in a previous by condensation work [12 ],of thebetween silsesquioxane four to eight oligomer silane molecul synthetizedes, with was a high a mixture number of compounds obtained by condensation of between four to eight silane molecules, with a high number ofintramolecular cycles and the complete hydrolysis of the alkoxide groups. The presence of hydrogen MaterialsMaterials2018 2017, 11, 11, 22, 22 6 of6 13 of 13 Materials 2017, 11, 22 6 of 13 of intramolecular cycles and the complete hydrolysis of the alkoxide groups. The presence of bondsofhydrogen intramolecular between bonds the carbonyl between cycles and groups the the carbonyl of complete the methacrylate groups hydrolysis of the and of methacrylate the the hydroxyls alkoxide and of groups. the the hydrolyzed hydroxyls The presence alkoxides of the of preventedhydrogenhydrolyzed the bonds condensationalkoxides between prevented reactionsthe carbonyl the from condensation groups concluding, of the reac so methacrylatetions open-cage from structuresand concluding, the hydroxyls of so the open polyhedral of- cage the oligomerhydrolyzedstructures silsesquioxanes of alkoxides the polyhedral were prevented obtained. oligomer the silsesquioxanesThese condensation open-cage reac were structurestions obtained. from were These concluding, selected open because- cage so open structures their-cage ashes showedstructureswere selected better of mechanical thebecause polyhedral their resistance ashes oligomer showed [12 ]. silsesquioxanes Next,better 5mechanical wt % of were this resistance open-cageobtained. [12 These]. oligomer Next, open 5 wt silsesquioxane- cage%. of structuresthis open was- incorporatedwerecage oligomerselected into because silsesquioxane the mixture their ashes ofwas unsaturated showed incorporated better polyester mechanicalinto the resinmixture resistance and of styrene unsaturated [12]. Next, prior 5polyester to wt polymerization %. of thisresin open and- to obtaincagestyrene theoligomer prior polyester/oligomer to silsesquioxane polymerization silsesquioxanewas to incorporated obtain the compositepolyester/oligomer into the mixture studied of below.silsesquioxane unsaturated polyestercomposite resin studied and below. styreneThe thermalprior to degradationpolymerization of to the obtain polyester/oligomer the polyester/oligomer silsesquioxane silsesquioxane composite compos wasite studied studied by The thermal degradation of the polyester/oligomer silsesquioxane composite was studied by the thebelow. same characterization techniques previously used for the polyester. Thermogravimetric analysis same characterization techniques previously used for the polyester. Thermogravimetric analysis (FigureThe8) showed thermal adegradation slight decrease of the inpolyester/oligomer both the decomposition silsesquioxane rate and composite weight was loss. studied The thermalby the (Figure 8) showed a slight decrease in both the decomposition rate and weight loss. The thermal degradationsame characterization processes were techniques delayed previously to higher used temperatures. for the polyester. For example, Thermogravimetric in the air atmosphere analysis at degradation processes were delayed to higher temperatures. For example, in the air atmosphere at 400(Figure◦C, the 8) thermostableshowed a slight polyester decrease exhibited in both the a weight decomposition loss of 90%, rate and whereas weight that loss. of theThe compositethermal degradation400 °C, the thermostable processes were polyester delayed exhibited to higher a temperatures. weight loss of For 90%, example, whereas in thatthe airof atmospherethe composite at material amounted only to 60%; the thermo-oxidative process that took place at 540 ◦C for the polyester 400material °C, the amoun thermostableted only polyester to 60%; the exhibited thermo -aoxidative weight loss process of 90%, that whereas took place that atof 540the °Ccomposite for the was observed at 600 ◦C for the composite material. materialpolyester amounwas observedted only at to600 60%; °C for the the thermo composite-oxidative material. process that took place at 540 °C for the polyester was observed at 600 °C for the composite material.

(a) (b)

Figure 8. Thermogravimetric(a) analysis (—) in air atmosphere and (---) in nitrogen,(b) at 15 °C/min for the Figure 8. Thermogravimetric analysis (—) in air atmosphere and (—) in nitrogen, at 15 ◦C/min for polyester and the polyester oligomeric silsesquioxane composite: (a) thermogravimetry (TG), and (b) theFigure polyester 8. Thermogravimetric and the polyester analysis oligomeric (—) silsesquioxanein air atmosphere composite: and (---) in (a nitrogen,) thermogravimetry at 15 °C/min (TG); for the and differential thermogravimetry (DTG). (b)polyester differential and thermogravimetry the polyester oligomeric (DTG). silsesquioxane composite: (a) thermogravimetry (TG), and (b) differential thermogravimetry (DTG). A reduction in the exothermicity of the thermal degradation reactions was also observed in the A reduction in the exothermicity of the thermal degradation reactions was also observed in the polyester/oligomerA reduction in silsesquioxanethe exothermicity composite of the thermal (Figure degradation 9) due to these reactions two effects: was also the observeddecrease in the polyester/oligomerpolyester/oligomerrate of the first reaction silsesquioxane silsesquioxane (which peaked composite at 365 (Figure (Figure°C), and 9)9) the due shift to to these these of the two two other effects: effects: degradation the the decrease decrease processes in inthe the ◦ raterate(those of of the withthe ﬁrst first peaks reaction reaction at 390 (which (which°C and peaked 540peaked °C) attoat 365higher365 °C),C), temperatures and the the shift shift (450 of of the°C the andother other 600 degradation degradation°C, respectively). processes processes ◦ ◦ ◦ ◦ (those(those withThe with volatile peaks peaks at species at 390 390 C °Cfound and and 540 in540 the °C)C) thermal toto higherhigher degradation temperatures of the (450 (450 polyester/oligomer °CC and and 600 600 °C,C, respectively). respectively).silsesquioxane compositeTheThe volatile volatile werespecies of species the same found found nature in in the theas thermalthose thermal found degradationdegradation in the decomposition of the polyester/oligomer of the thermostable silsesquioxane silsesquioxane polyester, compositecompositewith only were slight were of ofchanges the the same same in nature relativenature asas intensities thosethose found with in in temperaturethe the decomposition decomposition (Figure of 10). of the the thermostable thermostable polyester, polyester, withwith only only slight slight changes changes in in relative relative intensities intensities withwith temperaturetemperature (Figure (Figure 10). 10).

Figure 9. Differential thermal analysis (DTA) of thermostable polyester and polyester/oligomer Figuresilsesquioxane 9. Differential composite, thermal in air analysisatmosphere (DTA) (—) ofand thermostable in nitrogen ( --- polyester) (at 15 °C/min). and polyester/oligomer Figure 9. Differential thermal analysis (DTA) of thermostable polyester and polyester/oligomer silsesquioxane composite, in air atmosphere (—) and in nitrogen (---) (at 15 °C/min). silsesquioxane composite, in air atmosphere (—) and in nitrogen (—) (at 15 ◦C/min).

Materials 2018, 11, 22 7 of 13 Materials 2017, 11, 22 7 of 13

(a) (b)

FigureFigure 10.10. TheThe evolution withwith thethe temperature ofof thethe infraredinfrared signalsignal ofof eacheach identiﬁedidentified volatilevolatile speciesspecies duringduring the the thermal thermal degradation degradation of of the the polyester/oligomer polyester/oligomer silsesquioxane composite (a) in in the the air air atmosphere;atmosphere, andand ((bb)) inin nitrogen.nitrogen.

2.3.2.3. MechanismMechanism forfor thethe ThermalThermal DegradationDegradation ofof thethe ThermostableThermostable PolyesterPolyester andand thethe CompositeComposite To evaluate the pyrolytic decomposition mechanism, both the identified volatile species and the To evaluate the pyrolytic decomposition mechanism, both the identiﬁed volatile species and information included in the literature were considered. Comparison with the decomposition the information included in the literature were considered. Comparison with the decomposition products of polystyrene and polyester allowed us to identify which part of the polymer (main chain products of polystyrene and polyester allowed us to identify which part of the polymer (main chain or or chain-interconnections) gave rise to which degradation product. chain-interconnections) gave rise to which degradation product.

2.3.1.2.3.1. AldehydeAldehyde GenerationGeneration TheThe mainmain decompositiondecomposition productsproducts of the polystyrenepolystyrene identiﬁedidentified in the literature were toluene, ethylbenzene,ethylbenzene, styrene,styrene, αα-methylbenzene,-methylbenzene, benzaldehyde,benzaldehyde, andand phenylacetaldehyde.phenylacetaldehyde. In this study, only thethe aldehydesaldehydes werewere identiﬁedidentified (see(see TableTable1 ).1). These These gases gases were were mainly mainly produced produced in in the the process process taking taking placeplace aroundaround 365365◦ °CC inin bothboth atmosphericatmospheric conditions.conditions. AndersonAnderson andand FreemanFreeman [[2727]] describeddescribed thethe chemicalchemical mechanismmechanism ofof benzaldehydebenzaldehyde formationformation byby thethe thermalthermal degradationdegradation ofof styrenatedstyrenated polyesterpolyester inin thethe presencepresence ofof oxygen.oxygen. TheseThese authorsauthors proposedproposed thethe attack attack of of the the oxygen oxygen molecule molecule onon thethe carbon carbon in in the the α α position position ofof thethe phenyl phenyl group, group, with with thethe formationformation ofof anan unstableunstable hydroperoxidehydroperoxide intermediate.intermediate. Then,Then, thethe bondsbonds rupturedruptured andand rearrangedrearranged byby transferringtransferring thethe hydrogenhydrogen atom atom of of thethe carboncarbon inin thethe ββ position toto the benzoyl radical. This mechanism waswas alsoalso appliedapplied toto thethe thermo-oxidativethermo-oxidative degradationdegradation ofof polystyrenepolystyrene inin thethe presencepresence ofof oxygenoxygen [[2828]] but,but, unfortunately,unfortunately, couldcould not be used to to explain explain the the formation formation of of those those aldehydes aldehydes when when no no oxygen oxygen is ispr present.esent. InIn thethe presentpresent study,study, bothboth aldehydesaldehydes werewere generatedgenerated simultaneously.simultaneously. EvansEvans etet al.al. [[2020]] explainedexplained thethe simultaneoussimultaneous generation generation of of these these aldehydes aldehydes by by the the presence presence of of oxygenated oxygenated free free radicals, radicals, formed formed in thein the initial initial radical radical steps. steps. The The addition addition of of the the radical radical to to the the carbon carbon in in the theα αor orβ βposition position toto thethe phenylphenyl groupgroup produced produced either either one one or or the the other other aldehyde aldehyde (Figure (Figure 11 11).). This This mechanism mechanism could could explain explain the the simultaneoussimultaneous presencepresence ofof bothboth aldehydes,aldehydes, notnot onlyonly inin air,air, butbut alsoalso inin thethe nitrogennitrogen atmosphere,atmosphere, forfor thethe thermalthermal decompositiondecomposition processprocess thatthat tooktook placeplace aroundaround 365365◦ °C.C.

Figure 11. Chemical mechanism proposed by Evans for the formation of both aldehydes in the presence of oxygenated free radicals [20].

Materials 2017, 11, 22 7 of 13

(a) (b)

Figure 10. The evolution with the temperature of the infrared signal of each identified volatile species during the thermal degradation of the polyester/oligomer silsesquioxane composite (a) in the air atmosphere, and (b) in nitrogen.

2.3. Mechanism for the Thermal Degradation of the Thermostable Polyester and the Composite To evaluate the pyrolytic decomposition mechanism, both the identified volatile species and the information included in the literature were considered. Comparison with the decomposition products of polystyrene and polyester allowed us to identify which part of the polymer (main chain or chain-interconnections) gave rise to which degradation product.

2.3.1. Aldehyde Generation The main decomposition products of the polystyrene identified in the literature were toluene, ethylbenzene, styrene, α-methylbenzene, benzaldehyde, and phenylacetaldehyde. In this study, only the aldehydes were identified (see Table 1). These gases were mainly produced in the process taking place around 365 °C in both atmospheric conditions. Anderson and Freeman [27] described the chemical mechanism of benzaldehyde formation by the thermal degradation of styrenated polyester in the presence of oxygen. These authors proposed the attack of the oxygen molecule on the carbon in the α position of the phenyl group, with the formation of an unstable hydroperoxide intermediate. Then, the bonds ruptured and rearranged by transferring the hydrogen atom of the carbon in the β position to the benzoyl radical. This mechanism was also applied to the thermo-oxidative degradation of polystyrene in the presence of oxygen [28] but, unfortunately, could not be used to explain the formation of those aldehydes when no oxygen is present. In the present study, both aldehydes were generated simultaneously. Evans et al. [20] explained the simultaneous generation of these aldehydes by the presence of oxygenated free radicals, formed in the initial radical steps. The addition of the radical to the carbon in the α or β position to the phenyl group produced either one or the other aldehyde (Figure 11). This mechanism could explain the Materialssimultaneous2018, 11 p, 22resence of both aldehydes, not only in air, but also in the nitrogen atmosphere, for8 of the 13 thermal decomposition process that took place around 365 °C.

Figure 11. 11. ChemicalChemical mechanism mechanism proposed proposed by Evans by Evans for the for formation the formation of both of aldehydes both aldehydes in the presence in the presence of oxygenated free radicals [20]. Materialsof oxygenated2017, 11, 22 free radicals [20]. 8 of 13

2.3.2. Phthalic Anhydride Formation PhthalicPhthalic anhydride was mainly produced in the decomposition process that occurred around ◦ 390 °C.C. Ravey Ravey [29 [29] ]explained explained the the formation formation of ofthis this molecule molecule through through a mechani a mechanismsm that started that started with withthe elimination the elimination of the of the phthalic phthalic anhydride anhydride directly directly from from the the main main chain chain of of the the polyester, polyester, and proceeded with the recombination of the two radicals to obtain an ether group. If If this this were were the the case, case, 3-phenyl-2-propenoic3-phenyl-2-propenoic acid acid or or even even diethylene diethylene glycol glycol should should have havebeen detected, been detected, in addition in addition to phthalic to anhydride.phthalic anhydride. As neither As of neither these compoundsof these compounds were found were in found the volatiles, in the volatiles, Ravey’s mechanismRavey’s mechanism was not consideredwas not considered for the thermal for the decompositionthermal decomposition of the composite. of the composite. The mechanism proposedproposed byby SivasamySivasamy [ 30[30]] was was based based on on a βa -eliminationβ-elimination process. process. Hydrogens Hydrogens in thein theβ positionβ position to to the the benzoyl benzoyl radical radical could could contribute contribute a a cyclic cyclic six-center six-center transitiontransition state.state. According to this mechanism, the acid would ﬁrstfirst be obtained, and then dehydration would occur, leading to the anhydride. However, However, as as no no acid volatile compounds were were detected by by FTIR spectroscopy, this mechanism was also discarded. Anderson and Freeman [27] proposed a different mechanism for the generation of phthalic anhydride in the thermal decomposition process of the polyester (Figure 1212).). The reaction started with the homolytic cleavage of the bond closer to the ester group and the the formation formation of of free free ra radicals.dicals. Later, reorganization of the bonds would produce phthalic anhydride anhydride and and hydroxyester hydroxyester compounds. compounds. Figure 1313 showsshows that that hydroxyl hydroxyl groups groups were were detected detected simultaneously simultaneously to phthalic to phthalic anhydride anhydride in the in FTIR the spectroscopicFTIR spectroscopic characterization, characterization, so the so Anderson the Anderson and Freeman and Freeman mechanism mechanism could explain could explain the second the pyrolysissecond pyrolysis stage of stage the thermal of the thermal decomposition decomposition process process of the polyester of the polyester main chain main that chain started that aroundstarted 390around◦C in390 both °C atmosphericin both atmospheric conditions. conditions. These FTIR These signals FTIR corresponding signals corresponding to the hydroxyls to the hydroxyls vibrations havevibrations been have shown been with sh expandedown with expanded axes in Figure axes 13 in, Figure but were 13, notbut includedwere not inincluded Figures in7 and Figures 10 due 7 and to their10 due low to value.their low value.

Figure 12. 12. MechanismMechanism proposed proposed by Anderson by Anderson and Freeman and Freeman to explain to explain the formation the formation of phthalic of phthalicanhydride anhydride [27]. [27].

(a) (b)

Figure 13. The evolution with the temperature of the infrared signal of the phthalic anhydride and the hydroxyls groups in the thermal decomposition of the polyester (a) in the air atmosphere, and (b) in the nitrogen atmosphere.

Materials 2017, 11, 22 8 of 13

2.3.2. Phthalic Anhydride Formation Phthalic anhydride was mainly produced in the decomposition process that occurred around 390 °C. Ravey [29] explained the formation of this molecule through a mechanism that started with the elimination of the phthalic anhydride directly from the main chain of the polyester, and proceeded with the recombination of the two radicals to obtain an ether group. If this were the case, 3-phenyl-2-propenoic acid or even diethylene glycol should have been detected, in addition to phthalic anhydride. As neither of these compounds were found in the volatiles, Ravey’s mechanism was not considered for the thermal decomposition of the composite. The mechanism proposed by Sivasamy [30] was based on a β-elimination process. Hydrogens in the β position to the benzoyl radical could contribute a cyclic six-center transition state. According to this mechanism, the acid would first be obtained, and then dehydration would occur, leading to the anhydride. However, as no acid volatile compounds were detected by FTIR spectroscopy, this mechanism was also discarded. Anderson and Freeman [27] proposed a different mechanism for the generation of phthalic anhydride in the thermal decomposition process of the polyester (Figure 12). The reaction started with the homolytic cleavage of the bond closer to the ester group and the formation of free radicals. Later, reorganization of the bonds would produce phthalic anhydride and hydroxyester compounds. Figure 13 shows that hydroxyl groups were detected simultaneously to phthalic anhydride in the FTIR spectroscopic characterization, so the Anderson and Freeman mechanism could explain the second pyrolysis stage of the thermal decomposition process of the polyester main chain that started around 390 °C in both atmospheric conditions. These FTIR signals corresponding to the hydroxyls vibrations have been shown with expanded axes in Figure 13, but were not included in Figures 7 and 10 due to their low value.

MaterialsFigure2018 12., 11 ,Mechanism 22 proposed by Anderson and Freeman to explain the formation of phthalic9 of 13 anhydride [27].

(a) (b)

Figure 13. The evolution with the temperature of the infrared signal of the phthalic anhydride and Figure 13. The evolution with the temperature of the infrared signal of the phthalic anhydride and the the hydroxyls groups in the thermal decomposition of the polyester (a) in the air atmosphere, and (b) hydroxyls groups in the thermal decomposition of the polyester (a) in the air atmosphere; and (b) in inthe the nitrogen nitrogen atmosphere. atmosphere.

2.3.3. Main Chain Scission Carbon dioxide was detected in the thermal decomposition of the studied polyester and of the composite in both atmospheric conditions. In the air atmosphere, a pyrolytic reaction was followed by the occurrence of the consecutive combustion of the volatile species, with carbon dioxide as the main product. However, in the nitrogen atmosphere, carbon dioxide must be generated directly by the pyrolysis reactions. Several studies have been found in the literature that describe the thermal decomposition process of polyester, with carbon dioxide as the main reaction product [31–35]. Bikiaris et al. [35] proposed that the thermal degradation of an aliphatic polyester in a nitrogen atmosphere proceeded through the decomposition of the carboxylic groups in terminal positions of the polyester chains. Pohl [31] studied the effect of the chemical structure on the thermal stability of polymers and concluded that the presence of ester groups in the polymer main chain considerably reduced the thermal stability of the polymer. Therefore, he proposed that the degradation process took place by breaking bonds closer to the ester groups and producing small volatile molecules. Zimmermann [33] also described the starting degradation steps of polyesters as random scissions of the polymer chain by the ester unions. Baudry [18] even described these scission products as alkyl or alkoxy radicals.

2.3.4. Summary Considering the species that are identiﬁed, the thermal decomposition of this thermostable polyester in the interval 300–500 ◦C can be described as follows. From a thermal point of view, the weakest points in the polyester material are the ester groups. In the studied thermostable polyester, there are two types of ester groups (Figure 14): those in the aliphatic chains that come from the maleic anhydride (marked with a circle), and those closer to the aromatic rings that come from the phthalic anhydride (marked in the ﬁgure with a square). The scission of the polymer chain by both ester unions generates reactive alkoxy radicals as the reaction product, as described by Pohl, Zimmermann, and Baudry. Materials 2017, 11, 22 9 of 13

2.3.4. Summary Considering the species that are identified, the thermal decomposition of this thermostable polyester in the interval 300–500 °C can be described as follows. From a thermal point of view, the weakest points in the polyester material are the ester groups. In the studied thermostable polyester, there are two types of ester groups (Figure 14): those in the aliphatic chains that come from the maleic anhydride (marked with a circle), and those closer to the aromatic rings that come from the phthalic anhydride (marked in the figure with a square). The scission of the polymer chain by both ester unions generates reactive alkoxy radicals as the reaction product, as described by Pohl, Zimmermann, and Baudry. The second type of ester groups are stabilized by their proximity to the aromatic ring, so that the ester groups in the aliphatic chains break at lower temperatures (365 °C), generating oxygenated free radicals. These reactive radicals, which are close to the styrened chain-interconnections, simultaneously produce benzaldehyde and phenylacetaldehyde by the radical mechanism postulated by Evans. MaterialsThe2018 ester, 11 ,groups 22 closer to the aromatic rings break at higher temperatures (390 °C), generating10 of 13 phthalic anhydride by the mechanism postulated by Anderson and Freeman.

FigureFigure 14. 14. ChemicalChemical structure structure of of the the studied studied polyester. polyester. The two different ester groups are marked.

InThe the second present type study, of ester those groups differences are stabilized in behavior by their between proximity the ester to thegroup aromatic were observed, ring, so that either the inester the groups infrared in analysis the aliphatic of the chains volatile break species at lower (Figures temperatures 7 and 10) (365or in◦ theC), generatingDTA (Figure oxygenated 9). In the case free ofradicals. the DTA These experiments reactive radicals, in air, the which original are close polyester to the styrenedshowed chain-interconnections,a single exothermic peak simultaneously that was the resultproduce of benzaldehyde two overlapping and phenylacetaldehydepyrolysis-combustion by processes the radical, corresponding mechanism postulated to both bytypes Evans. of ester groups.The This ester single groups exothermic closer to thepeak aromatic split in two rings for break the composite, at higher temperatures due to a stabilization (390 ◦C), of generating the ester phthalic anhydride by the mechanism postulated by Anderson and Freeman. In the present study, those differences in behavior between the ester group were observed, either in the infrared analysis of the volatile species (Figures7 and 10) or in the DTA (Figure9). In the case of the DTA experiments in air, the original polyester showed a single exothermic peak that was the result of two overlapping pyrolysis-combustion processes, corresponding to both types of ester groups. This single exothermic peak split in two for the composite, due to a stabilization of the ester groups marked with a square (by interaction with the oligomer) that shifted the second pyrolysis-combustion process 25 ◦C to higher temperatures. Such an increase in the temperature of thermal decomposition of a polydimethylsiloxane (PDMS) polymer by a reaction with a close-cage POSS oligomer with vinyl groups was reported in the literature by Yang et al. [36]. Only when these silsesquioxane oligomers were chemically incorporated was the thermal stability improved. In this work, a certain stabilization in the ester group of the aliphatic chains was also observed for the composite. Although their pyrolysis process occurred at the same temperature as that of the polyester sample (365 ◦C), the slight decrease in the heat released indicated there was a slowdown of the decomposition rate for the composite. A similar decrease in the heat released was found in the second process. The two effects resulted in a lower thermal degradation of the composite.

3. Materials and Methods

3.1. Materials 3-Methacryloxypropyltrimethoxysilane (MAPTMS) supplied by Sigma-Aldrich Química, (Madrid, Spain) S.A., was selected as the precursor of the silsesquioxane oligomer, for it has reactive unsaturations in the methacrylate group. Unsaturated polyester resin (obtained by reaction between maleic anhydride, phthalic anhydride, and ethylene glycol, and later diluted in styrene) was kindly supplied by Ferro Spain, (Almassora, Castellon, Spain) S.A. Accelerator NL-51P, a cobalt(II) 2-ethylhexanoate, 6 wt % Co in solvent, was employed as an accelerator, and Butanox LA, methyl ethyl ketone peroxide (MEKP) solution, as an initiator. Both additives were supplied by AzkoNobel (Barcelona, Spain).

3.2. Preparation of the Silsesquioxane Oligomer Synthesis was performed under acid-catalyzed conditions. In a typical synthesis, 20 g (0.08 mole) of MAPTMS were poured into a 200 mL container, and 6.5 g (0.36 mole) of acidiﬁed water (HCl 0.01 M) Materials 2018, 11, 22 11 of 13

was slowly added while the silane was stirred. Then, the synthesis was performed with a H2O:Si mole ratio = 4.5:1 with constantly stirring for 264 h. The characterization of the obtained product is described in detail in a previous paper [12].

3.3. Preparation of the Thermostable Polyester Unsaturated polyester resin was polymerized with 1 wt % of the accelerator NL-51P and 2 wt % of the initiator Butanox LA. The methyl ethyl ketone peroxide (MEKP) was selected as the initiator instead of benzoyl peroxide owing to the fact that the polyesters obtained were more stable [37]. The mixture was poured into molds before gel time to obtain samples with the dimensions 20 mm × 80 mm × 3 mm.

3.4. Preparation of the Polyester/Oligomer Silsesquioxane Composite The polyester/oligomer silsesquioxane composite was prepared as the polyester, but by adding 5 wt % of silsesquioxane oligomer prior to adding the initiator and pouring into the molds. Samples of similar dimensions to the polyester were obtained.

3.5. Sample Characterization The Fourier transform infrared spectroscopy (FTIR) spectra of the samples were obtained with a NICOLET 6700 THERMO spectrometer (Thermo Fischer Scientiﬁc Inc., Waltham, MA, USA), at an average of 16 scans at a resolution of 4 cm−1. The thermal gravimetric analysis (TGA) and differential thermal analysis (DTA) were performed on a TGA/SDTA 851e (Mettler Toledo International Inc., L’Hospitalet de Llobregat, Spain), by placing cylindrical samples (5 mm diameter, 0.5 mm thickness) in the platinum crucible and heating up to 700 ◦C at a heating rate of 15 ◦C/min, in both ﬂowing nitrogen and static air atmospheres. The differential scanning calorimetry (DSC) measurements were performed on a STA 449 C JUPITER (Netzsch, Sant Joan Despí, Barcelona, Spain) instrument. The volatiles were characterized by infrared spectroscopy with a coupled TGA-IR to the DSC instrument.

4. Conclusions Polyester was prepared by the polymerization of an unsaturated ester resin with styrene. An open-cage silsesquioxane with methacrylate groups was employed in the preparation of the polyester/oligomer silsesquioxane composite to improve the thermal behavior of the polymer. Thermogravimetric analysis combined with infrared spectroscopy has been shown to be a powerful technique in the thermal decomposition studies of polymers, even though previous knowledge of the original chemical structure of the polymeric system is needed to propose a pyrolysis mechanism. The thermal degradation of the thermostable polyester and that of the polyester/oligomer silsesquioxane composite are complex, heterogeneous processes of several reactions that depend on temperature and atmosphere. Between 300 ◦C and 500 ◦C, two groups of chemical transformations were observed, with scission of the polymer chains by both ester unions. These processes likely resulted in the generation of reactive radicals, as the literature describes. The scission of the polymer chain by the ester groups in the aliphatic chains occurred around 365 ◦C, and those reactive radicals close to the styrened chain-interconnections simultaneously generated benzaldehyde and phenylacetaldehyde by the radical mechanism postulated by Evans et al. [20]. The scission of the polymer chain by the ester groups closer to the aromatic rings, which was observed around 390 ◦C for the polyester, produced phthalic anhydride by the mechanism postulated by Anderson and Freeman [27]. The presence of 5 wt % open-cage oligomeric silsesquioxanes in the polyester composite affected both pyrolytic transformations: there was a slight decrease in the decomposition rate of the ﬁrst process at 365 ◦C, and a displacement of the second pyrolytic process to higher temperatures; this second effect was due to a stabilization of the ester groups closer to the aromatic rings. Both effects contributed Materials 2018, 11, 22 12 of 13 to a more gradual release of heat in the combustion reactions, thus decreasing the rate of thermal degradation and improving the ﬁre resistance of the polyester composite.

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. Bruins, P.F. Unsaturated Polyester Technology; Gordon and Breach: New York, NY, USA, 1976. 2. Laoutid, F.; Bonnaud, L.; Alexandre, M.; López-Cuesta, J.-M.; Dubois, P. New prospects in flame retardant polymer materials: From fundamentals to nanocomposites. Mater. Sci. Eng. 2009, 63, 100–125. [CrossRef] 3. Camino, G.; Delobel, R. Fire Retardancy for Polymeric Materials; Grand, A.F., Wilkie, C.A., Eds.; Marcel Dekker: New York, NY, USA, 2000. 4. 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] 5. Franchini, E.; Galy, J.; Gérard, J.-F.; Tabuani, D.; Medici, A. Influence of POSS structure on the fire retardant properties of epoxy hybrid networks. Polym. Degrad. Stab. 2009, 94, 1728–1736. [CrossRef] 6. Zhang, W.; Li, X.; Jiang, Y.; Yang, R. Investigations of epoxy resins flame-retarded by phenyl silsesquioxanes of cage and ladder structures. Polym. Degrad. Stab. 2013, 98, 246–254. [CrossRef] 7. Li, Z.; Yang, R. Flame retardancy, thermal and mechanical properties of sulfonate-containing polyhedral oligomeric silsesquioxane (S-POSS)/polycarbonate composites. Polym. Degrad. Stab. 2015, 116, 81–87. [CrossRef] 8. Song, L.; He, Q.; Hu, Y.; Chen, H.; Liu, L. Study on thermal degradation and combustion behaviors of PC/POSS hybrids. Polym. Degrad. Stab. 2008, 93, 627–639. [CrossRef] 9. Blanco, I.; Bottino, F.A. The influence of the nature of POSSs cage’s periphery on the thermal stability of a series of new bridged POSS/PS nanocomposites. Polym. Degrad. Stab. 2015, 121, 180–186. [CrossRef] 10. Lewicki, J.P.; Pielichowski, K.; Jancia, M.; Hebda, E.; Albo, R.L.F.; Maxwell, R.S. Degradative and morphological characterization of POSS modified nanohybrid polyurethane elastomers. Polym. Degrad. Stab. 2014, 104, 50–56. [CrossRef] 11. Carvalho, H.W.P.; Suzana, A.F.; Santilli, C.V.; Pulcinelli, S.H. Structure and thermal behavior of PMMA-polysilsesquioxane organic-inorganic hybrids. Polym. Degrad. Stab. 2014, 104, 112–119. [CrossRef] 12. Bautista, Y.; Gozalbo, A.; Mestre, S.; Sanz, V. Improvements in char strength with an open cage silsesquioxane flame retardant. Materials 2017, 10, 567. [CrossRef][PubMed] 13. Sanz, S.V.; Bautista, R.Y.; Silva, M.G. Thermostable Copolymer Materials Having Flame-Retardant Behaviour, And Objects Manufactured From Such Materials. WO2010070175 A1, 2010. 14. Moldoveanu, S.C. Analytical Pyrolysis of Synthetic Organic Polymers; Elsevier: Amsterdam, The Netherlands, 2005. 15. Evans, S.J.; Haines, P.J.; Skinner, G.A. The effects of structure on the thermal degradation of polyester resins. Thermochim. Acta 1996, 278, 77–89. [CrossRef] 16. Evans, S.J.; Haines, P.J.; Skinner, G.A. The thermal degradation of polyester resins II. The effects of cure and fillers on degradation. Thermochim. Acta 1997, 291, 43–49. [CrossRef] 17. Budrugeac, P.; Segal, E. Thermooxidativa degradation of an unsatured polyester resin. J. Therm. Anal. Calorim. 1997, 49, 183–191. [CrossRef] 18. Baudry, A.; Dufay, J.; Reignier, N.; Mortaigne, B. Thermal degradation and fire behaviour of unsaturated polyester with chain ends modified by dicyclopentadiene. Polym. Degrad. Stab. 1998, 61, 441–452. [CrossRef] 19. Agrawal, J.P.; Sarwade, D.B.; Makashir, P.S.; Mahajan, R.R.; Dendage, P.S. Thermal degradation studies of novel diethylene glycol based unsaturated polyesters in air. Polym. Degrad. Stab. 1998, 62, 9–14. [CrossRef] 20. Evans, S.J.; Haines, P.J.; Skinner, G.A. Pyrolysis-gas chromatographic study of a series of polyester thermosets. J. Anal. Appl. Pyrolysis 2000, 55, 13–28. [CrossRef] 21. Kicko-Walczak, E. Kinetics of thermal decomposition of unsaturated polyester resins with reduced flammability. J. Appl. Polym. Sci. 2003, 88, 2851–2857. [CrossRef] Materials 2018, 11, 22 13 of 13

22. Shih, Y.F.; Jeng, R.J.; Wei, K.M. Carbon black containing interpenetrating polymer networks based on unsatured polyester/epoxy III: Thermal and pyrolysis analysis. J. Anal. Appl. Pyrolysis 2003, 70, 129–144. [CrossRef] 23. Shih, Y.F.; Jeng, R.J. Thermal degradation behaviour and kinetic analysis of unsaturated polyester-based composite and IPNs by conventional and modulated thermogravimetric analysis. Polym. Degrad. Stab. 2006, 91, 823–831. [CrossRef] 24. Kandare, E.; Kandola, B.K.; Price, D.; Nazaré, S.; Horrocks, R.A. Study of the thermal decomposition of flame-retarded unsaturated polyester resins by thermogravimetric analysis and Py-GC/MS. Polym. Degrad. Stab. 2008, 93, 1996–2006. [CrossRef] 25. Tibiletti, L.; Longuet, C.; Ferry, L.; Coutelen, P.; Mas, A.; Robin, J.J.; Lopez-Cuesta, J.M. Thermal degradation and fire behaviour of unsaturates polyesters filled with metallic oxides. Polym. Degrad. Stab. 2011, 96, 67–75. [CrossRef] 26. Mark, J.E. Physical Properties Polymers Handbook; American Institute of Physics: Woodbury, NY, USA, 1996. 27. Anderson, D.A.; Freeman, E.S. The kinetics of the thermal degradation of the synthetic styrenated polyester, laminac 4116. J. Appl. Polym. Sci. 1959, 1, 192–199. [CrossRef] 28. Singh, B.; Sharma, N. Mechanistic implications of plastic degradation. Polym. Degrad. Stab. 2008, 93, 561–584. [CrossRef] 29. Ravey, M. Pyrolysis of unsaturated polyester resin. Quantitative aspects. J. Polym. Sci. Polym. Chem. 1983, 21, 1–5. [CrossRef] 30. Sivasamy, P.; Palaniandavar, M.; Vijayakumar, C.T.; Lederer, K. The role of β-hydrogen in the degradation of polyesters. Polym. Degrad. Stab. 1992, 38, 15–21. [CrossRef] 31. Pohl, H.A. The Thermal Degradation of Polyesters. J. Am. Chem. Soc. 1951, 73, 5660–5661. [CrossRef] 32. Buxbaum, L.H. The Degradation of Poly(ethylene terephthalate). Angew. Chem. Int. Ed. 1968, 7, 182–190. [CrossRef] 33. Zimmermann, H. Developments in Polymer Degradation; Grassie, N., Ed.; Applied Science: London, UK, 1984; Volume 5. 34. Persenaire, O.; Alexandre, M.; Degée, P.; Dubois, P. Mechanisms and Kinetics of Thermal Degradation of Poly(ε-caprolactone). Biomacromolecules 2001, 2, 288–294. [CrossRef][PubMed] 35. Bikiaris, D.N.; Chrissafis, K.; Paraskevopoulos, K.M.; Triantafyllidis, K.S.; Antonakou, E.V. Investigation of thermal degradation mechanism of an aliphatic polyester using pyrolysis-gas chromatography-mass spectrometry and a kinetic study of the effect of the amount of polymerisation catalyst. Polym. Degrad. Stab. 2007, 92, 525–536. [CrossRef] 36. Yang, D.; Zhang, W.; Yao, R.; Jiang, B. Thermal stability enhancement mechanism of poly(dimethylsiloxane) composite by incorporating octavinyl polyhedral oligomeric silsesquioxanes. Polym. Degrad. Stab. 2013, 98, 109–114. [CrossRef] 37. Bansal, R.K.; Mittal, J.; Singh, P. Thermal stability and degradation studies of polyester resins. J. Appl. Polym. Sci. 1989, 37, 1901–1908. [CrossRef]

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