Recent Advances in Bio-Based Flame Retardant Additives for Synthetic Polymeric Materials

polymers

Review Recent Advances in Bio-Based Flame Retardant Additives for Synthetic Polymeric Materials

Christopher E. Hobbs

Department of Chemistry, Sam Houston State University, Huntsville, TX 77340, USA; [email protected]; Tel.: +1-936-294-3750 Received: 28 December 2018; Accepted: 24 January 2019; Published: 31 January 2019

Abstract: It would be difficult to imagine how modern life across the globe would operate in the absence of synthetic polymers. Although these materials (mostly in the form of plastics) have revolutionized our daily lives, there are consequences to their use, one of these being their high levels of flammability. For this reason, research into the development of flame retardant (FR) additives for these materials is of tremendous importance. However, many of the FRs prepared are problematic due to their negative impacts on human health and the environment. Furthermore, their preparations are neither green nor sustainable since they require typical organic synthetic processes that rely on fossil fuels. Because of this, the need to develop more sustainable and non-toxic options is vital. Many research groups have turned their attention to preparing new bio-based FR additives for synthetic polymers. This review explores some of the recent examples made in this field.

Keywords: ﬂame retardant; bio-based; green chemistry

1. Introduction Although instilling flame resistance has been a subject of interest since the ancient Egyptians [1], “modern” chemists have only been on the case since the 18th and 19th centuries, as evidenced by the words of American poet Emily Dickinson [2]: “Ashes denote that fire was—revere the grayest pile, for the departed creature’s sake, that hovered there awhile—fire exists the first in light, and then consolidates, only the chemist can disclose, into what carbonates.” The same era saw the publication of landmark work by French chemist Joseph Louis Gay-Lussac (who studied flame resistance with the intent of protecting theatres) and Scottish chemist M. M. Pattison Muir, who published The Chemistry of Fire in 1893, only a few years after Emily Dickinson’s death [3–5]. In the century-and-a-half since, our understanding of the molecular world, as it pertains to fire, has progressed significantly and the scientific community has made tremendous advances so that chemists can fully disclose into what carbonates. The 20th century saw enormous advances in the chemical sciences. In particular, the rise of polymer chemistry led to a revolution in daily life across the industrialized world. Because of this, these synthetic materials (e.g., plastics and fibers) have become ubiquitous. However, the extremely high levels of flammability often exhibited by these materials is a consequence of their daily use. But, the advances made in polymer chemistry led to an increase in our understanding of fire resistance and provided chemists the motivation to study and develop better flame resistant (FR) polymers. Early commercial examples include: Bakelite, Nomex, Kevlar, and others. [3] This is understandably an important topic of industrial and academic research considering that structure fires cause $6.7 billion worth of damage, 12,300 injuries, and >2500 deaths annually in the United States alone [6]. The massive California wild fires in the summer and autumn of 2018 were a stark reminder of the threat that fires pose. For these reasons, the number of reports detailing the preparation of new FR polymeric materials has exploded in recent years (Figure1).

Polymers 2019, 11, 224; doi:10.3390/polym11020224 www.mdpi.com/journal/polymers Polymers 2018, 10, x FOR PEER REVIEW 2 of 30

threat that fires pose. For these reasons, the number of reports detailing the preparation of new FR Polymers 2019, 11, 224 2 of 31 polymeric materials has exploded in recent years (Figure 1).

Figure 1.1. GraphGraph representing representing the the number number of peer-reviewed of peer-revie journalwed journal articles (foundarticles in (found SciFinder) in SciFinder) describing flamedescribing resistant flame (FR) resistant polymers (FR) per polymers year between per year 1969 between and 2017. 1969 and 2017. This growth in the number of reports of new FR materials has been spurred by advances in science This growth in the number of reports of new FR materials has been spurred by advances in as well as closer scrutiny by governmental regulatory bodies, many of which have sought to decrease science as well as closer scrutiny by governmental regulatory bodies, many of which have sought to and outright ban the use of halogenated FR materials in the last few decades [7]. In light of this, decrease and outright ban the use of halogenated FR materials in the last few decades [7]. In light of it has been the responsibility of the scientific community to develop new, safer alternatives to these this, it has been the responsibility of the scientific community to develop new, safer alternatives to traditional FR systems, of which phosphorus-containing species are some of the most studied [8]. To do these traditional FR systems, of which phosphorus-containing species are some of the most studied so, it is important to understand how polymeric materials burn: combustion of polymeric material [8]. To do so, it is important to understand how polymeric materials burn: combustion of polymeric involves evolution of combustible volatiles through decomposition in an oxygen-rich atmosphere. material involves evolution of combustible volatiles through decomposition in an oxygen-rich No matter the type of FR used, the mechanisms for flame retardancy occur by one of two routes (or a atmosphere. No matter the type of FR used, the mechanisms for flame retardancy occur by one of combination of the two): (i) condensed phase or (ii) vapor phase [9]. Flame resistance/retardancy in two routes (or a combination of the two): (i) condensed phase or (ii) vapor phase [9]. Flame the condensed phase often involves the formation of char layer that can stop the spread of the flame resistance/retardancy in the condensed phase often involves the formation of char layer that can stop and self-extinguish, protecting the material by limiting the transfer of heat and oxygen to the polymer. the spread of the flame and self-extinguish, protecting the material by limiting the transfer of heat Vapor phase flame resistance/retardancy involves terminating the radical chemistry involved in fire, and oxygen to the polymer. Vapor phase flame resistance/retardancy involves terminating the radical often the mechanism by which traditional halogenated FR materials operate (Table1). chemistry involved in fire, often the mechanism by which traditional halogenated FR materials operate (Table Table1). 1. Mechanisms of flame retardancy based on phase and example of each.

PhaseTable 1. Mechanisms of flame retardancy Method based on phase and example of Example each.

CondensedPhase formation ofMethod protective char Example P-containing FR VaporCondensed termination of free formation radicals of protective involved char in combustion P-containing halogenated FR FR Vapor termination of free radicals involved in combustion halogenated FR The chemical community has undergone an immense shift in our approach toward synthesis in recentThe years. chemical This hascommunity been catalyzed has undergone by our increased an immens cognizancee shift in our of the approach environmental toward synthesis impact that in ourrecent reliance years. onThis petroleum has been hascatalyzed had and by the our role incr iteased has playedcognizance in global of the climate environmental change. Inimpact addition, that our reliance current hydrocarbonon petroleum feedstockshas had and are the renewable role it has only played on ain geologicalglobal climate timescale. change. So, In addition, chemists recognizeour current and hydrocarbon understand feedstocks the need forare morerenewable renewable only on and a sustainablegeological timescale. sources. ThisSo, chemists “Green Chemistry”recognize and revolution understand has impactedthe needevery for more aspect re ofnewable chemistry, and from sustainable biochemistry sources. to synthesisThis “Green [10]. TheChemistry” development revolution of FRs has is impacted no different. every In aspect fact, oneof chemistry, of the biggest from recentbiochemistry objectives to synthesis has been [10]. the synthesisThe development and use ofof “bio-based”FRs is no different. FR materials, In fact, which one areof morethe biggest renewable recent and objectives sustainable has since been they the aresynthesis not generated and use of from “bio-based” our finite FR petroleum materials, reserves which are [11 more]. Although renewable there and are sustainable hazards in since working they withare not any generated chemical (especiallyfrom our finite FRs thatpetroleum can decompose reserves into[11]. toxic Although species there upon are combustion), hazards in the working use of morewith any renewable chemical (bio-based) (especially resources FRs that serves can decompose as a step in into the toxic right species direction upon toward combustion), our realization the use of renderingof more renewable the science (bio-based) more sustainable. resources In serves general, as a flame step in resistance the right can direction be introduced toward intoour realization polymeric materialsof rendering through the science the incorporation more sustainable. of low molecular In general, weight flame additives, resistance either can throughbe introduced blending into or covalentpolymeric attachment. materials Thethrough past severalthe incorporation years has witnessed of low molecular great growth weight in this additives, field and either many differentthrough bio-based materials have been used. The purpose of this review is to summarize recent advances made Polymers 2018, 10, x FOR PEER REVIEW 3 of 30 blending or covalent attachment. The past several years has witnessed great growth in this field and many different bio-based materials have been used. The purpose of this review is to summarize Polymers 2019, 11, 224 3 of 31 recent advances made in this field. This review is organized around the bio-based material used and will mainly be focused on recent reports detailing the incorporation of such material into synthetic polymericin this field. materials. This review is organized around the bio-based material used and will mainly be focused on recent reports detailing the incorporation of such material into synthetic polymeric materials. 2. Bio-Based FR Materials 2. Bio-Based FR Materials 2.1. Tannic Acid and Related Materials 2.1. Tannic Acid and Related Materials Tannic acid is a member of the class of polyphenols, which are naturally occurring materials containing,Tannic as acid the is name a member suggests, of the multiple class ofphenolic polyphenols, moieties which in their are molecular naturally occurringstructure (as materials can be seencontaining, by the structure as the name of tannic suggests, acid multiple in Figure phenolic 2). Some moieties of the most in their abundant molecular natural structure polyphenols (as can are be theseen tannins by the (both structure condensed of tannic and acid hydrolysable), in Figure2). wh Someich are of thefound most all abundantthroughout natural the plant polyphenols kingdom. Tanninsare the tanninsare the (bothfourth condensed most abundant and hydrolysable), material available which from are plants found [12] all throughoutand are utilized the plant for antibacterialkingdom. Tannins and insect are the repellent fourth most properties abundant [13] material. Despite available their frombiological plants prevalence, [12] and are utilizedthe use forof tanninsantibacterial as additives and insect for repellent synthetic properties polymers [13 is]. Despiterelatively their limited. biological This prevalence, is possibly the due use to of color tannins or incompatibilityas additives for syntheticwith hydrophobic polymers polymers is relatively [14]. limited. However, This tannins is possibly do exhibit due to colorexcellent or incompatibility char-forming withability hydrophobic (due to the polymersformation [of14 ].cross-linked However, tanninsaromatic do structures exhibit excellent from the char-forming inner galloyl ability units (dueduring to combustion)the formation as of well cross-linked as antioxidant aromatic properties; structures thes frome characteristics the inner galloyl make units tannins during attractive combustion) targets as wellfor bio-based as antioxidant FR materials. properties; In fact, these these characteristics properties allow make them tannins to attractiveprovide natural targets fire for protection bio-based for FR treesmaterials. [15]. In fact, these properties allow them to provide natural fire protection for trees [15].

Figure 2.2. MolecularMolecular structures structures of of a condenseda condensed tannin tannin monomer monomer (left )( andleft) aand hydrolysable a hydrolysable tannin (tannictannin (tannicacid, right acid,). right).

Taking inspirationinspiration from from Mother Mother Nature, Nature, Tributsch Tribut andsch Fiechterand Fiechter [15] showed [15] showed that naturally-occurring that naturally- occurringtannins could tannins be usedcould to be instill used synthetic to instill synthetic polymers polymers with higher with levels higher of ﬂamelevels retardancy.of flame retardancy. To show Tothis, show the authors this, the acquired authors tanninsacquired from tannins the barks from ofthe quebracho, barks of quebracho, Canary Island Canary pine Island (Pinus pine canariensis (Pinus), canariensisand giant sequoia), and giant (Sequoiadendron sequoia (Sequoiadendron) trees, by extraction) trees, by into extraction a water/acetone into a water/acetone mixture using mixture sonication. using Thesonication. obtained The material obtained was material then mixed was withthen amixed synthetic with polymer a synthetic (acrylonitrile-butadiene-styrene polymer (acrylonitrile-butadiene- (ABS) styreneresin), in (ABS) which resin), the thermal in which properties the thermal and properti combustiones and performancecombustion perfor of barmance specimens of bar were specimens tested wereand compared tested and to compared various tree to barks.various Thermogravimetric tree barks. Thermogravimetric analysis (TGA) analysis under (TGA) argon revealedunder argon that ◦ revealedthe tree barksthat the displayed tree barks relatively displayed high relatively levels of high thermostability levels of thermostability up to 600 C,up leaving to 600 °C, behind leaving up behindto 60% carbonizedup to 60% carbonized solid. On thesolid. other On the hand, other ABS hand, resin ABS underwent resin underwent one step one decomposition step decomposition around ◦ around400 C to400 leave °C to behind leave ca. behind 2% charred ca. 2% residue. charred Theresi thermaldue. The properties thermal properties of the ABS of could the ABS be affected could bybe affectedmixing with by mixing tannin, with however. tannin, Mixing howeve ABSr. withMixing either ABS 30 with or 50 eitherw/w% 30 tannin or 50 resultedw/w% tannin in materials resulted that in materialsunderwent that slightly underwent more complexslightly more decomposition complex decomposition (similar to tree (similar barks) whileto tree leaving barks) behindwhile leaving higher behindlevels of higher charred levels remains—up of charred to remains—up 27%. Similar to results 27%. wereSimilar noticed results when were ABS noticed resin when was mixed ABS resin with various amounts of Canary Island pine bark instead of extracted tannin. The authors note that the inclusion of tannin increased the limiting oxygen index (LOI), but provided no values. Polymers 2018, 10, x FOR PEER REVIEW 4 of 30

Polymers 2019, 11, 224 4 of 31 was mixed with various amounts of Canary Island pine bark instead of extracted tannin. The authors note that the inclusion of tannin increased the limiting oxygen index (LOI), but provided no values. AA few years later, Celzard and coworkers [[13]13] showed that condensed tannin-formaldehyde polymeric foamsfoams couldcould be be used used as as superior superior ﬂame flame retardants retardants when when compared compared to typical to typical phenolic phenolic foams. foams.Additionally, Additionally, such materials such materials are more are attractive more attractive since tannins since tannins are relatively are relatively nontoxic nontoxic and bio-based. and bio- based.They explained They explained that the that superior the supe FR naturerior FR is nature due to is the due stable to the char stable formation. char formation. The materials The materials described describedshowed more showed difﬁcult more ignitions difficult and ignitions lower heatand releaselower heat rates release as well. rates The as tannin-based well. The tannin-based foams tested foamsshowed tested heat showed release ratesheat release per unit rates area per (HRRPUA) unit area of(HRRPUA) 12 kW/m of2, while12 kW/m pure2, while phenolic pure and phenolic epoxy andfoams epoxy peaked foams at 106peaked and at 314 106 kW/m and 3142. Further, kW/m2. ignitionFurther, timesignition of tannintimes of foams tannin exceeded foams exceeded 100 s, while 100 s,phenolic while phenolic foams were foams 6 s. were 6 s. InIn 2015, 2015, the the laboratories laboratories of of Wang Wang and and Schiraldi Schiraldi [16] [16 reported] reported on the on incorporation the incorporation of tannic of tannic acid intoacid aerogels into aerogels made made of sodium of sodium montmorillonite montmorillonite clay clay and and epoxy epoxy polymers. polymers. Advantageously, Advantageously, this preparation utilized relatively “green” procedures that involved mixing clay with 1,4-butanediol diglycidyldiglycidyl ether ether 1 1andand triethylenetetramine triethylenetetramine 2 (as2 a(as curing a curing agent) agent) and subjecting and subjecting the mixture the mixture to freeze- to dryingfreeze-drying in water in (Figure water (Figure 3). Incorporation3). Incorporation of tannic of tannic acid could acid could also be also carried be carried out by out freeze-drying by freeze-drying the samethe same mixture mixture with with ca. 2% ca. tannic 2% tannic acid. acid. Further Further incorp incorporationoration of tannic of tannic acid acid could could be accomplished be accomplished by coatingby coating the theaerogels aerogels in inwater water or orethanol ethanol solutions solutions of ofeither either 2% 2% tannic tannic acid acid or or 2% 2% tannic tannic acid-0.2% acid-0.2% FerricFerric complexes.

FigureFigure 3. 3. MolecularMolecular structures structures of of 1,4-butanediol 1,4-butanediol diglycidyl diglycidyl ether ether 11 and triethylenetetramine 2..

TheThe FR properties of these materials were testedtested and described.described. The The authors authors found found that that all aerogelsaerogels exhibited exhibited some some levels levels of of self-extinguishing behavior behavior when when exposed exposed to to a a propane propane torch torch for for 3030 s. Cone Cone calorimetry calorimetry provided provided peak peak heat heat release release rate rate (PHRR), (PHRR), time time to to ignition ignition (TTI), (TTI), total heat releaserelease (THR), (THR), time time to to peak peak of of heat heat release release rate rate (TTPHRR), (TTPHRR), and and fire fire growth growth rate rate (FIGRA). (FIGRA). The The results results areare summarized summarized in in Table Table 22 (in (in which which the the samples samples are arenamed named based based on the on percentage the percentage content content of epoxy of andepoxy clay, and for clay, example, for example, a composite a composite of 20% of epoxy, 20% epoxy, 5% clay, 5% and clay, 2% and tannic 2% tannic acid is acid E20C5T2). is E20C5T2). The 2 authorsThe authors note note that thattypical typical PHRR PHRR of ofepoxy epoxy resins resins is isca. ca. 1200 1200 kW/m kW/m2 whilewhile the the clay clay composite composite (even (even 2 without the incorporation of tannic acid) exhibitedexhibited valuesvalues nono higherhigher thanthan 400400 kW/mkW/m2.. Incorporation Incorporation ofof tannic tannic acid acid lowered lowered this value value by by up up to to 20%. 20%. In In addition addition to to the the flammabili flammabilityty properties, properties, the the authors authors describedescribe the the differences differences in mechanical properties. Th Thisis is is an an issue issue that that may often be overlooked, but isis important important nonetheless; nonetheless; this this is is especially especially true true if if such such materials materials have have a a future future in in commercialization. commercialization. The authors noted that E20C5 aeroge aerogelsls were highly flexible flexible and could recover ca. 95% 95% of of the the original original shape.shape. The The incorporation incorporation of of only only 2% 2% tannic tannic acid, acid, however, however, resulted resulted in in a a much much more more dense and and stiff foam.foam. Furthermore, eveneven though though flame flame retardancy retardancy could could be enhancedbe enhanced by coating by coating the samples the samples in tannic in acid/FeCl solutions, this resulted in a highly colored foam, which may not be aesthetically pleasing tannic acid/FeCl3 3 solutions, this resulted in a highly colored foam, which may not be aesthetically pleasingin a commercial in a commercial sense. sense. Table 2. FR data for epoxy-clay composites. Table 2. FR data for epoxy-clay composites. Entry TTI 1 (s) PHRR 2 (kW/m2) THR 3 TTPHRR 4 FIGRA 5 Residue (%) LOI 6 (%) Entry TTI 1 (s) PHRR 2 (kW/m2) THR 3 TTPHRR 4 FIGRA 5 Residue (%) LOI 6 (%) E20C5 6 407 47 160 2.5 18.5 19 E20C5 6 407 47 160 2.5 18.5 19 E20C5 coated 3 335 61 175 1.9 22.5 21 E20C5E20C5T2 coated 7 3 370 335 5161 175 165 1.9 2.2 22.5 21.1 21 20 E20C5T2E20C5T2 coated 37 319 370 6251 165 185 2.2 1.7 21.1 21.6 20 21 E20C5T21 time coated to ignition; 2 peak 3 heat release 319rate; 3 total heat release; 62 4 time 185 to peak of heat 1.7 release rate; 21.65 fire growth rate; 216 1limiting time to oxygen ignition; index. 2 peak heat release rate; 3 total heat release; 4 time to peak of heat release rate; 5 fire growth rate; 6 limiting oxygen index . Moustafa and coworkers [17] showed that the FR properties of tannic acid-ABS composites could be improvedMoustafa by and the additioncoworkers of CaCO[17] showed3 (extracted that fromthe FR seashell properties wastes) of as tanni a bio-filler.c acid-ABS Cone composites calorimetry couldshowed be thatimproved ABS containing by the addition 5% tannic of CaCO acid displayed3 (extracted aTTI from of seashell 22 s and wastes) PHRR of as 696 a bio-filler. kW/m2, whileCone

Polymers 2019, 11, 224 5 of 31 the incorporation of 25% seashell waste into the composite exhibited a TTI value of 46 s and a 45% decrease in PHRR. Though the extent to which the amount of tannic acid effected the FR properties was not tested. Inorganic hydroxides have been used as FR additives for a while now [18]. They are advantageous since they are relatively non-toxic and smokeless. However, polymer composites of metal hydroxides can exhibit poor adhesion to surfaces. Taking inspiration from mussels (in which catechol and amino groups are important for adhesion), it was recently shown that this problem could be addressed in a synergistic manner, by the incorporation of tannic acid-Fe(III) complexes and Mg(OH)2 into a polyamide-6 (PA 6, an industrially-important polymer) matrix [19]. Tannic acid-Fe complexes can be used to coat materials on the surface because of their strong chelating properties and were used in this study to stabilize polymer–hydroxide composites through non-covalent interactions. The composites were prepared by single-screw extrusion and studied using a variety of techniques. The authors showed that the incorporation of Mg(OH)2 into the PA 6 matrix could increase the LOI from 22.1% (0 wt % Mg(OH)2) to 29.1% (45 wt % Mg(OH)2). This could be improved with the addition of tannic acid–Fe complexes. For instance, the incorporation of 45 wt % Fe-complex and hydroxide led to an LOI of more than 31%. The authors note that in the absence of tannic acid–Fe complex, composites modified with 30 and 45 wt % hydroxide exhibited the same UL-94 classification. They concluded that there must exist a synergy between the metal hydroxide and tannic acid–Fe complex. Through TGA it was proposed that Mg(OH)2 and MgO (from decomposed magnesium hydroxide), along with Fe, facilitated the decomposition of tannic acid to form a protective char. Cone calorimetry was employed to further study this synergy. Unmodified PA 6 exhibited a PHRR of 512 kW/m2. This value could be reduced by 26% through the incorporation of Mg(OH)2 and tannic acid-Fe. Further, PA 6 left behind no residue while the Mg(OH)2/tannic acid-Fe composite left 32.4 wt % residue. Considering that ca. 80% of deaths attributed to fires are actually caused by smoke inhalation [20], the reduction of smoke formation is an important aspect of FR. The authors studied the smoke production rate (SPR) and total smoke production (TSP) and found that the addition of 45 wt % 2 −1 Mg(OH)2 and tannic acid–Fe to the PA 6 matrix reduced the SPR from ca. 0.225 to 0.177 m s 2 and reduced the TSP by 38% (1.307 m ). Notably, the inclusion of tannic acid–Fe(III) and Mg(OH)2 suppressed the formation of toxic gases as well. It has been shown that the thermal decomposition of tannic acid commences with the outer galloyl moieties to form pyrogallol and CO2 [21]. Mosurkal and Nagarajan [12] hypothesized that chemical modification and stabilization of the outer galloyl units would lead to enhanced FR properties of tannic acid. This was tested through reaction of tannic acid with terephthaloyl chloride 3 to form a complex network (tannic acid-terephthalate, TAT, 4) that could be used as a FR coating for nylon fabric (Figure4). First, the authors conducted a series of experiments aimed at deducing whether or not the modified tannic acid 4 exhibited enhanced thermal stability compared to tannic acid. The TGA analysis showed unmodified tannic acid to have a T5% (temperature at which 5% mass loss occurred) ◦ ◦ to be 185 C and a Tmax (temperature at which the maximum mass loss occurred) to be 333 C, while the values for 4 were 299 and 531 ◦C, respectively. Additionally, the char yield of tannic acid was ca. 27% while 4 was slightly higher at ca. 37%. Upon completion of these studies, the authors prepared nylon samples treated with 4 by submerging nylon fabric in an aqueous suspension of 4 and acetic acid at 100 ◦C. The fabric was then washed, dried, and subjected to a modified ASTM D6413 vertical flame test. This test allows for the measurement of burning rate (determined by the height the flame reached within 10 s), afterflame time (i.e., amount of time recorded after ignition source is removed), and char length (distance between bottom of sample and furthest point the flame reached). The data are summarized in Table3. As can be seen, nylon 66 coated with 4 showed enhanced FR properties when compared to nylon 66, both uncoated and coated with tannic acid. The two latter samples burned completely (127 mm was the length of the sample) and exhibited burning rates essentially double that of the 4-treated samples. Furthermore, the 4-treated nylon exhibited no melt dripping. Samples coated with tannic acid exhibited Polymers 2019, 11, 224 6 of 31 properties almost identical to that of the control. The authors note that this is because of the high-water solubilityPolymers 2018 of, 10 tannic, x FOR acid, PEER whichREVIEW was washed out of the sample before the flame tests. 6 of 30

Figure 4. PreparationPreparation of tannic acid-terephthalate (TAT) 44..

Upon completion of these studies,Table 3.theVertical authors flame prepared test data. nylon samples treated with 4 by submerging nylon fabric in an aqueous suspension of 4 and acetic acid at 100 °C. The fabric was then Sample Burning Rate (mm/s) Afterflame (s) Char Length (mm) washed, dried, and subjected to a modified ASTM D6413 vertical flame test. This test allows for the measurement Nylonof burning 66 rate (determined by 12.5 the height the flame 18 reached within 10127 s), afterflame Nylon 66 (coated with tannic acid) 10.2 25 127 time (i.e., amount of time recorded after ignition source is removed), and char length (distance Nylon 66 (coated with 4) 6.4 0 76 between bottom of sample and furthest point the flame reached). The data are summarized in Table 3. As can be seen, nylon 66 coated with 4 showed enhanced FR properties when compared to nylon 66, bothPolydopamine uncoated and (PDA) coated is another with tannic attractive acid. polyphenolic The two latter material samples and burned is environmentally completely (127 friendly mm andwas non-toxic.the length Ellisonof the sample) et al. [22 and] reported exhibited on theburning use of rates mussel-inspired essentially double PDA-treated that of polyurethanethe 4-treated (PU)samples. foams Furthermore, recently. Furthermore, the 4-treated catechols nylon exhibited have long no been melt used dripping. as radical Samples scavengers, coated an with important tannic mechanismacid exhibited of actionproperties for FR almost materials. identical The PU-dopamine to that of the compositecontrol. The foams authors were note prepared that this by immersingis because PUof the foam high-water in a buffered solubility solution of tannic of dopamine acid, which hydrochloride. was washed A series out of of the thermal sample analyses before the and flame flame tests. tests were then conducted. The TGA revealed that untreated PU (control) quickly reduced in weight and left ◦ no observable residue when heated toTable 600 3.C, Vertical while flame pure polydopaminetest data. had almost 64 wt % residue at 600 ◦C. The PHRR and THR of control PU foam were found to be 498.3 W/g and 22.7 kJ/g, respectively. Sample Burning Rate (mm/s) Afterflame (s) Char Length (mm) The PDA showed undetectable PHRR and a THR of 2.4 kJ/g. All PU foams were subjected to butane Nylon 66 12.5 18 127 torch flameNylon for 66 (coated 10 s to with qualitatively tannic acid) study flammability. 10.2 The control PU 25 foam immediately 127 ignited and formed a significantNylon 66 (coated amount with of 4) PU melt pool; stopping6.4 this melt drip is0 important to reduce 76 the spread of flame. The PDA-coated foams exhibited no melt dripping. The amount of PDA did improve the FR properties.Polydopamine The PDA1D (PDA) (PDA is coatedanother for attractive 1 day, 5.3 wtpolyphenolic % PDA) underwent material significantand is environmentally damage upon exposurefriendly and to the non-toxic. flame while Ellison PDA3D et (PDAal. [22] coated reported for 3 days,on the 15.9 use wt %of PDA)mussel-inspired greatly retained PDA-treated its shape andpolyurethane self-extinguished, (PU) foams protecting recently. a sizableFurthermore, portion cate of thechols sample have fromlong been the flame. used Coneas radical calorimetry scavengers, was employedan important to furthermechanism analyze of theaction foams. for UntreatedFR materials. PU foamThe PU-dopamine showed a PHRR composite value of 734foams kW/m were2. Incorporationprepared by immersing of 15.9 wt %PU PDA foam reduced in a buffered this value solution to 239 kW/m of dopami2. Additionally,ne hydrochloride. the authors A series showed of thatthermal the generationanalyses and of flame toxic gasestests were (i.e., COthen and conducte CO2) wered. The reduced TGA revealed from 0.053 that kg/kg untreated and 2.792PU (control) kg/kg, respectively,quickly reduced in the in control weight foam and to left 0.013 no and observable 0.028 kg/kg. residue when heated to 600 °C, while pure polydopamine had almost 64 wt % residue at 600 °C. The PHRR and THR of control PU foam were found to be 498.3 W/g and 22.7 kJ/g, respectively. The PDA showed undetectable PHRR and a THR of 2.4 kJ/g. All PU foams were subjected to butane torch flame for 10 s to qualitatively study flammability. The control PU foam immediately ignited and formed a significant amount of PU melt pool; stopping this melt drip is important to reduce the spread of flame. The PDA-coated foams exhibited no melt dripping. The amount of PDA did improve the FR properties. The PDA1D (PDA coated for 1 day, 5.3 wt % PDA) underwent significant damage upon exposure to the flame while PDA3D (PDA coated for 3 days, 15.9 wt % PDA) greatly retained its shape and self-extinguished,

Polymers 2018, 10, x FOR PEER REVIEW 7 of 30 protecting a sizable portion of the sample from the flame. Cone calorimetry was employed to further analyze the foams. Untreated PU foam showed a PHRR value of 734 kW/m2. Incorporation of 15.9 wt % PDA reduced this value to 239 kW/m2. Additionally, the authors showed that the generation of toxic gases (i.e., CO and CO2) were reduced from 0.053 kg/kg and 2.792 kg/kg, respectively, in the Polymerscontrol 2019foam, 11 to, 224 0.013 and 0.028 kg/kg. 7 of 31

2.2. Phytic Acid 2.2. Phytic Acid As mentioned earlier, one of the most vigorous areas of FR research has been the development of non-halogenatedAs mentioned earlier, flame one retardants. of the most The vigorous reasoning areas ofbehind FR research this hashas beenbeen the spurred development by the of non-halogenatedenvironmental dangers ﬂame retardants.that halogenated The reasoning organics behind can pose this to has the been environment. spurred by Furthermore, the environmental many dangersgovernment that regulatory halogenated bodies organics have can imposed pose to thestrict environment. regulations Furthermore,on their use. manySome governmentof the best regulatoryperforming bodiesreplacements have imposedare phosph strictorus regulations containing onmaterials, their use. either Some in polymeric of the best form performing or as low replacementsmolecular weight are phosphorus additives to containing synthetic materials, or natural either polymers in polymeric [8]. One form of orthe as main low molecular mechanisms weight of additivesaction exhibited to synthetic by phosphorus or natural polymersFR is their [8 excellent]. One of ability the main to mechanismsform protective of action char layers exhibited which by phosphorusaffects self-extinguishing FR is their excellent and protects ability the to sample form protective from thechar spread layers of the which flame. affects There self-extinguishing exists a plethora andexamples protects of phosphorus the sample from FR in the the spread literature, of the many ﬂame. of Therewhich exists are outside a plethora the scope examples of this of phosphorusreview and FRthe inauthor the literature, would encourage many of which the reader are outside to explore the scope recent of reviews this review [23–28]. and the author would encourage the readerAn issue to explore at hand recent is the reviews fact that [23 many–28]. low molecular weight phosphorus species can be toxic. AlthoughAn issue their at toxicity hand is could the fact presumably that many be low reduce moleculard through weight covalent phosphorus attachment species to a can polymer, be toxic. it Althoughmay still be their undesirable toxicity to could work presumably with such hazardous be reduced species. through One covalent safe, non-toxic attachment alternative to a polymer,is phytic itacid may 5, stillthe bestructure undesirable of which to work is withshown such in hazardous Figure 5. species.Phytic Oneacid safe,5 is non-toxica naturally alternative occurring is phyticpolyphosphate acid 5, the ester structure of inositol of whichand is an is shownimportant in Figurestorage5. form Phytic of phosphorus acid 5 is a in naturally plant tissues occurring [29]. polyphosphateSeveral research ester groups of inositol have taken and isadvantage an important of th storageis compound form of as phosphorus a natural FR in additive. plant tissues In 2014, [29]. SeveralZhang and research coworkers groups [30] have described taken advantage the preparatio of thisn of compound a polyelectrolyte as a natural complex FR additive.(PEC) composed In 2014, Zhangof a naturally and coworkers occurring [30 polymer,] described chitosan the preparation 6, which is of cationic a polyelectrolyte under dilute complex acidic (PEC) conditions composed (the ofstructure a naturally of which occurring is shown polymer, in Figure chitosan 5), and6 anionic, which 5 is as cationic a FR treatment under dilute for the acidic synthetic conditions copolymer (the structureof ethylene-vinyl of which acetate is shown (EVA), in Figure an industrially-imp5), and anionic 5ortantas a FR material treatment [30]. for The the PECs synthetic were copolymerprepared by of ethylene-vinylthe addition of acetate aqueou (EVA),s solutions an industrially-important of anionic 5 into aqueous material solutions [30]. The of PECs cationic were 6 prepared. The resulting by the additionmaterial was of aqueous incorporated solutions into of EVA anionic composites5 into aqueous (0, 5, 10, solutions and 20 wt of cationic% PEC)6 by. The melt resulting blending material at 110 ◦ was°C. incorporated into EVA composites (0, 5, 10, and 20 wt % PEC) by melt blending at 110 C.

Figure 5. MolecularMolecular structures structures of ethylene-vinyl acetate (EVA), 5,, andand 66..

T T The authors ﬁrstfirst testedtested thethe thermalthermal propertiesproperties ofof6 6and andthe thePEC PEC by by TGA. TGA. The The T5%5% andand Tmaxmax for 6 ◦ (control) were 271 and 298 °C,C, respectively. Chitosan 6 left behind char layer residue of 43 wt % at ◦ T ◦ 600 °C.C. The PEC exhibited earlierearlier decomposition, butbut thethe Tmaxmax waswas 17 °CC higher higher and the char residue ◦ at 600 °CC was was 56 56 wt wt %. %. This This suggested suggested enhancement enhancement of ofthe the thermal thermal stability. stability. The The TGA TGA analysis analysis of the of theEVA–PEC EVA–PEC composites composites revealed revealed more more complex complex degr degradationadation patterns patterns for for composites composites with with higher content of PEC. Additionally, the higher the PEC content,content, the higher the residual mass (0 wt % for EVA and 12% for EVA containingcontaining 2020 wtwt % PEC). Furthermore,Furthermore, PHRR decreased as the content of PEC increased (801(801 and 552 W/g for for EVA EVA and and EVA EVA containing 20 wt % PEC, respecti respectively).vely). In turn, the LOI values also increased: 20 and 22.9% for EVA and EVA containing 20 wt % PEC, respectively. During the UL-94 vertical burn test, only EVA composites of 20 wt % PEC reached V-2 rating. The same group applied a similar PEC system based on 5 and polyethylenimine 7 (Figure6) as FR treatment for polypropylene (PP), one of the most widely used and important performance polymers [31]. The authors prepared PP composites of 5, 10, and 20 wt % PEC and compared their Polymers 2018,, 10,, xx FORFOR PEERPEER REVIEWREVIEW 8 of 30 values also increased: 20 and 22.9% for EVA and EVA containing 20 wt % PEC, respectively. During the UL-94 vertical burn test, only EVA composites of 20 wt % PEC reached V-2 rating. The same group applied a similar PEC system based on 5 and polyethylenimine 7 (Figure 6) as PolymersThe2019 same, 11, 224group applied a similar PEC system based on 5 and polyethylenimine 7 (Figure8 6) of as 31 FR treatment for polypropylene (PP), one of the most widely used and important performance polymers [31]. The authors prepared PP composites of 5, 10, and 20 wt % PEC and compared their thermal stability to control PP, 55,, 77,, andand compositescomposites ofof PPPP withwith thethe latterlatter two.two. The TGA (summarized 5% in TableTable4 4)) revealed revealed a a reduction reduction in inT T5%5%with with the the incorporation incorporation of of PEC PEC (due (due to to the the earlier earlier degradation degradation of max ofPEC). PEC). The TheTmax Tmaxvalues values increased increased as didas did char char formation formation for PP–PECfor PP–PEC composite. composite. The The authors authors note note that thatthis isthis due is todue the to enhanced the enhanced thermo-oxidative thermo-oxidative stability stability of PEC of PEC over over PP alone. PP alone.

Figure 6. Molecular structure of 7.7.

Table 4. Thermal analysis of polypropylene-polyelectrolyte (PP–PEC) composites. Table 4. Thermal analysis of polypropylene-polyelectrolyte (PP–PEC) composites. ◦ ◦ Sample T5% (°C) Tmax (°C) Residue (%) Sample T5% T( 5%C) (°C) TmaxT(maxC) (°C) Residue Residue (%) (%) 7 247 345 4 77 247247 345345 4 4 5 253 308 78 55 253253 308308 78 78 PEC 241 321 61 PEC 241241 321321 61 61 PP 271 354 1 PP 271271 354354 1 1 PP-5% PEC 263 357 3 PP-5% PEC PEC 263 263 357 357 3 3 PP-10% PEC 256 368 6 PP-10%PP-10% PEC PEC 256 256 368 368 6 6 PP-20% PEC 263 379 11 PP-20%PP-20% PEC PEC 263 263 379 379 11 11 PP-20% 7 287 392 1 PP-20% 77 287 392 1 PP-20% 5 287260 392321 1 10 PP-20% 55 260260 321321 10 10

The authors next tested the flammability properties of the samples using combustion calorimetry.The authors It was next revealed tested that the ﬂammability the incorporatio propertiesn of PEC of into the samplesthe PP matrix using could combustion reduce calorimetry. the PHRR fromIt was 1129 revealed to as thatlow as the 815 incorporation W/g (PP-20% of PEC). PEC into This the illustrated PP matrix the could synergistic reduce therelationship PHRR from between 1129 to 5 andas low 7, as PP 815 composites W/g (PP-20% with PEC). 5 or This7 alone illustrated exhibited the PHRR synergistic values relationship of 938 and between965 W/g,5 respectively.and 7, as PP Furthermore,composites with LOI5 valuesor 7 alone were exhibited measured PHRR and values a noticeable of 938 andincrease 965 W/g,was observed respectively. between Furthermore, pristine PP,LOI PP-5% values PEC, were PP-10% measured PEC, and and a noticeable PP-20% PEC increase (18.0%, was 19.7%, observed 22.0%, between and 25.1%, pristine respectively). PP, PP-5% PEC, PP-10%In order PEC, andto be PP-20% considered PEC (18.0%,an intumescent 19.7%, 22.0%, flame and retardant 25.1%, respectively).(IFR), the material must contain an acid source,In order a to blowing be considered agent, and an intumescent a charring agent. ﬂame All retardant of which (IFR), are the important material for must the containformation an acidof a densesource, layer a blowing of char agent, foam and upon a charring combustion agent. All[32]. of A which common are important IFR used for is the based formation on ammonium of a dense polyphosphatelayer of char foam 8, uponpentaerythritol combustion 9, [ 32and].A melamine common IFR10 (Figure used is based7). With on ammoniumthe interest polyphosphate of green and sustainable8, pentaerythritol chemistry9, and in melaminemind, Wang10 (Figureet al. [33]7). reported With the on interest the reaction of green between and sustainable bio-based chemistry 5 and 10 toin mind,form a Wang synergistic et al. [33 IFR] reported additive on for the reactionPP. The betweenproposed bio-based structure5 and(suggested10 to form from a synergistic very limited IFR spectroscopicadditive for PP. techniques) The proposed of the structure resulting (suggested product from11 is veryshown limited in Figure spectroscopic 8. Product techniques) 11 was used of theas, both,resulting an acid product source11 isand shown blowing in Figure agent8 and. Product was used11 was in conjunction used as, both, with an dipentaerythritol acid source and blowing 12 as a charringagent and agent. wasused in conjunction with dipentaerythritol 12 as a charring agent.

Figure 7. Chemical structures of 8, 9, and 10.

To gain insight into the thermal properties of these materials, the authors ﬁrst employed TGA for the analysis of 11, 12, and a mixture of the two. Dipentaerythritol 12 underwent a one-step Polymers 2018, 10, x FOR PEER REVIEW 9 of 30

Figure 7. Chemical structures of 8, 9, and 10. Polymers 2019, 11, 224 9 of 31

decomposition, beginning at 250 ◦C, to leave no residue. 11, on the other hand, showed a more complex, three-step decomposition beginning with 15% weight loss between 250 and 350 ◦C, representative of release of ammonia and water. Dehydration followed between 350 and 600 ◦C to produce water and poly(phytic acid). Lastly, poly(phytic acid) decomposition occurs, generating phosphorus oxide (above 600 ◦C) to leave 25 wt % residue at 800 ◦C. Composites composed of PP were prepared by melt-mixing in a twin-screw extruder at 190 ◦C followed by hot-pressing under 10 MPa for 15 min. The LOI values of various formulations were measured and were found to increase upon increasing content of 11 and 12. Control PP exhibited a value of 17% while the sample containing 7.5 and 22.5% 12 11 and , respectively, exhibited an LOI of 28.5%. Furthermore, the latter obtained a V-0 rating in the UL-94 vertical burn test whereas the control PP exhibited extensive combustion and melt drip. Figure 8. Proposed structure of 11 and structure of 12. The TGA experiments provided insight into the thermal properties of the composites. Unsurprisingly, control PP decomposed completely leaving behind no residue at 500, 600, and 700 ◦C. This could be PolymersTo 2018gain, 10 insight, x FOR PEERinto theREVIEW thermal properties of these materials, the authors first employed TGA9 of 30 increased, albeit slightly, to 6.16, 4.87, and 3.58%, respectively, upon the increase of 12 (7.5 wt %) and for the analysis of 11, 12, and a mixture of the two. Dipentaerythritol 12 underwent a one-step 11 (22.5 wt %) content. Figure 7. Chemical structures of 8, 9, and 10. decomposition, beginning at 250 °C, to leave no residue. 11, on the other hand, showed a more complex, three-step decomposition beginning with 15% weight loss between 250 and 350 °C, representative of release of ammonia and water. Dehydration followed between 350 and 600 °C to produce water and poly(phytic acid). Lastly, poly(phytic acid) decomposition occurs, generating phosphorus oxide (above 600 °C) to leave 25 wt % residue at 800 °C. Composites composed of PP were prepared by melt-mixing in a twin-screw extruder at 190 °C followed by hot-pressing under 10 MPa for 15 min. The LOI values of various formulations were measured and were found to increase upon increasing content of 11 and 12. Control PP exhibited a value of 17% while the sample containing 7.5 and 22.5% 12 and 11, respectively, exhibited an LOI of 28.5%. Furthermore, the latter obtained a V-0 rating in the UL-94 vertical burn test whereas the control PP exhibited extensive combustion and melt drip. The TGA experiments provided insight into the thermal properties of the composites. Unsurprisingly, control PP decomposed completely leaving behind no residue at 500, 600, and 700 °C. This could be increased, albeit slightly, to 6.16, 4.87, and 3.58%, respectively, upon Figure 8. Proposed structure of 11 and structure ofof 1212.. the increase of 12 (7.5 wt %) and 11 (22.5 wt %) content. AlongAlongTo gain withwith insight other other into sub-disciplines sub-disciplines the thermal ofproperties chemistry,of chemistry, of the these ideas the materials, andideas principles andthe authorsprinciples underlying first underlyingemployed the philosophy TGAthe philosophyoffor green the analysis chemistry of green of have 11chemistry, 12 had, and a have dramatic a mixture had a impact dramatic of the on two. theimpact fieldDipentaerythritol on of the polymer field of chemistry. polymer 12 underwent chemistry. [34] In a particular, one-step [34] In particular,thedecomposition, utilization the ofutilization beginning more sustainable ofat more250 resources°C, sustainable to leave for developingno resources residue. newfor11, materialsdevelopingon the other or newnew hand, methodsmaterials showed to or preparea morenew methodsexistingcomplex, materialsto three-step prepare is existing ofdecomposition tremendous materials importance. beginningis of tremendous with Considering 15%importance. weight the toll Consideringloss that between society the 250 has toll takenand that 350society on the°C, hasenvironment,representative taken on the the of environment, preparationrelease of ammonia of the biodegradable preparation and water. polymersof bi Dehydrationodegradable is also important followedpolymers [ between35is ].also One important of350 the and most [35].600 widely °C One to ofusedproduce the biodegradablemost water widely and used polymerspoly(phytic biodegradable is poly(lacticacid). polymersLastly, acid) poly(phytic (PLA,is poly(lactic the structureacid) acid) decomposition (PLA, of which the isstructure shownoccurs, in ofgenerating Figurewhich9 is). shownPoly(lacticphosphorus in Figure acid) oxide 9). can (abovePoly(lactic be prepared 600 acid)°C) from to can leave vegetation be prepar25 wt ed% (as residuefrom opposed vegetation at 800 to petroleum) °C. (as Composites opposed and to breaks composedpetroleum) down of and into PP breakslacticwere prepared acid, down an into environmentally by lactic melt-mixing acid, an benigninenvironmentally a twin-screw product [extruder35 benign]. atproduct 190 °C [35]. followed by hot-pressing under 10 MPa for 15 min. The LOI values of various formulations were measured and were found to increase upon increasing content of 11 and 12. Control PP exhibited a value of 17% while the sample containing 7.5 and 22.5% 12 and 11, respectively, exhibited an LOI of 28.5%. Furthermore, the latter obtained a V-0 rating in the UL-94 vertical burn test whereas the control PP exhibited extensive combustion and melt drip. The TGA experiments provided insight into the thermal properties of the composites. Unsurprisingly, control PP decomposed completely leaving behind no residue at 500, 600, and 700 °C. This could be increased, albeit slightly, to 6.16, 4.87, and 3.58%, respectively, upon FigureFigure 9. 9. StructureStructure of of poly(lactic poly(lactic acid) acid) (PLA). (PLA). the increase of 12 (7.5 wt %) and 11 (22.5 wt %) content. AlthoughAlthoughAlong with more more other sustainable, sustainable, sub-disciplines PLA PLA suffers suffers of chemistry, from from the the same samethe detriments detrimentsideas and as asprinciples many many other other underlying plastics, plastics, it itthe is is highlyhighlyphilosophy flammable flammable of green (much (much chemistry work work havehas has gone gonehad ainto into dramatic the the development development impact on the of of fieldFR FR PLA PLA of polymer [36], [36], some some chemistry. of of the studies [34] In utilizingutilizingparticular, bio-based bio-based the utilization FR FR for for PLA PLAof willmore will be sustainable bedescribed described throughout resources throughout thisfor thisreview).developing review). To solve new To solve thismaterials problem, this problem, or Tang new andTangmethods coworkers and to coworkers prepare [37] usedexisting [37] used 5 as materi 5a asFR a alscoating FR is coating of tremendousfor forPLA PLA non-woven non-wovenimportance. fabrics. fabrics. Considering The The fabrics fabrics the weretoll were that prepared prepared society usingusinghas taken melt-blownmelt-blown on the environment, spinningspinning technology technology the preparation and and were were of treated bi odegradabletreated with with5 simply polymers5 simply by immersion isby also immersion important into a PAinto [35]. solution a OnePA (containingof the most 50,widely 100, 150,used 200 biodegradable or 250 g of 5 )polymers for ten minutes, is poly(lactic followed acid) by (PLA, drying. the The structure flammability of which of the is dry,shown non-woven in Figure fabrics 9). Poly(lactic were then acid) tested. can be A correlationprepared from of LOI vegetation with mass (as gain opposed (essentially to petroleum) the amount and breaks down into lactic acid, an environmentally benign product [35].

Figure 9. Structure of poly(lactic acid) (PLA).

Although more sustainable, PLA suffers from the same detriments as many other plastics, it is highly flammable (much work has gone into the development of FR PLA [36], some of the studies utilizing bio-based FR for PLA will be described throughout this review). To solve this problem, Tang and coworkers [37] used 5 as a FR coating for PLA non-woven fabrics. The fabrics were prepared using melt-blown spinning technology and were treated with 5 simply by immersion into a PA

Polymers 2019, 11, 224 10 of 31 of PLA absorbed) of each sample was observed. The LOI values of up to 36% were measured, for samples treated with 250 g of 5 (untreated PLA exhibited LOI of 26.3%). Vertical burn test revealed that none of the treated samples exhibited afterflame time and showed smaller burn area than the control. There was also a noticeable downward trend in PHRR, HRC, and THR between the control (465 W/g, 461 J/gK, and 19 kJ/g, respectively), PLA treated with 100 g of 5 (328 W/g, 327 J/gK, and 14 kJ/g, respectively), and PLA treated with 250 g of 5 (280 W/g, 278 J/gK, and 12 kJ/g, respectively). ◦ Temperature of maximum heat release (Tmax) also showed this trend for three samples, 375, 371, 365 C, respectively. The TGA data was also provided and importantly showed an increase in the char residue of these three samples. While the control left behind no residue, PLA treated with 100 and 250 g of 5 left behind ca. 6 and 16% residue at 600 ◦C, respectively. Although an improvement over untreated PLA fabric, 5 on its own left 45% char residue at 600 ◦C. Further, the authors noted that the treated fabrics in this study were not durable as a few washes with water were all that were needed to remove most of the 5 from the samples. A few years later Laoutid et al. [38] reported on the use of a combination of 5 and lignin as route for instilling FR properties to PLA. Lignin is a structurally-complex, highly crosslinked, naturally occurring polymer found in wood and bark. It has been used many times as a filler in polymer matrices to instill some FR properties, which it can do thanks to it aromatic structure that aids in char formation (FR polymeric materials using this bio-based material are outlined later in this review). The authors prepared polymer composites through melt processing PLA, 5, and lignin in various proportions. The lignin samples used were prepared using either the kraft process or organosolv [39]. The authors first studied the thermal stability of the compositions as well as the additives alone. They noted that the two lignin samples exhibited different thermal behaviors, the details of which are outside the scope of this review. Suffice it to say, all additives exhibited char formation upon degradation. Combinations of lignin and 5 exhibited thermal decompositions completely different than the additives alone: a mixture composed of 50% lignin and 50% 5 exhibited a T5% value much lower than any one of the components alone. This is possibly due to some dehydration processes that release low molecular weight components at lower temperatures. These mixtures also left char residues of ca. 50% at 800 ◦C. Next, the authors tested the thermal stabilities of PLA composites with lignin, PA, and both lignin ◦ and 5. The PLA alone exhibited a one-step decomposition in TGA (Tmax = 371 C) to leave no residue. The incorporation of lignin resulted in lower decomposition temperatures (20 wt % kaft and 20 wt % ◦ organosolv exhibited Tmax values of 332 and 271 C, respectively) and higher char residue formation (5 and 11%, respectively). Phytic acid 5 exhibited similar results. Mixtures of lignin and 5, however, resulted in TGA curves that closely resemble that of pristine PLA: PLA composite of 10% 5 and ◦ 10% organosolv lignin revealed Tmax of 381 C, in which a synergistic effect of the PA and lignin are responsible. Cone calorimetry was employed to study the FR properties with a heat flux of 35 kW/m2. The results are summarized in Table5. As can be seen, pristine PLA readily ignites and burns completely to leave no residue. Incorporation of lignin seemed to increase the ignitability of PLA. Similar effects have been observed before and could be attributable to increase in heat absorption and earlier release of combustible products. Although the ignitability increased, the PHRR and THR both significantly decreased. The use of 5 as an additive alone had less of an effect on the TTI, but also led to decreases in PHRR and THR. Both lignin samples and PA led to higher char residue than the control PLA. Using PA and lignin in concert led to a TTI closer to that of PLA. Phytic acid 5 helps to reduce the thermo-degradation of PLA by lignin, as was suggested by TGA. Although the use of 5 alone did seem to improve the FR properties of PLA, the authors note that it is not ideal as it can leach from the polymer matrix within days. The use of PA in conjunction with lignin seemed to improve this. This synergistic effect manifested itself in scanning electron microscopy (SEM) as it seemed that the 5 helped in lignin particle dispersion within the PLA matrix. It was noted that this may ensure a more homogeneous formation of char upon combustion. Further analysis using UL-94 tests revealed that Polymers 20192018, 1110, 224x FOR PEER REVIEW 11 of 30 31 the PLA–PA composite as well as composite of PLA, 5, and lignin samples gave V-2 ratings while the controlthe PLA–PA (PLA) composite and PLA–lignin as well compos as compositeites gave of PLA, no classification.5, and lignin samples gave V-2 ratings while the controlLastly, (PLA) although and PLA–lignin not directly composites related to gave the FR no pr classification.operties (but vastly important nonetheless), the authors stated that the incorporation of 5 and lignin had detrimental effects on the mechanical properties. TensileTable strength 5. Cone calorimetry for PLA was data measured and UL-94 resultsat 70.3 forMPa. PLA– For5–lignin example, composites. incorporation of PA and Kraft ligninEntry separately and together TTI (s) decreasedPHRR (kW/m this2 )value THR to 27.7, (MJ/m 16.9,2) andResidue 35.8 (%)MPa, respectively. UL-94 Elongation at break,PLA as well as impact 87 resistance, 390 also decreased. 90 0 NC 20% 5 71 270 74 13 V-2 20%Table Kraft 5. Cone calorimetry 37 data and UL-94 310 results for PLA– 715–lignin composites. 17 NC 1 20% Organosolv 26 260 67 19 NC 15%Entry5/5% KraftTTI 77 (s) PHRR 285 (kW/m2) THR 74 (MJ/m2) Residue 13 (%) V-2 UL-94 10%PLA5/10% Kraft87 43 220390 7490 120 V-2 NC 15% 520%/5% 5 Organosolv71 61 250270 78 74 13 13 V-2 V-2 10%20%5/10% Kraft Organosolv 37 46 250 310 65 71 13 17 V-2 NC 1 20% Organosolv 26 1 No classification. 260 67 19 NC 15% 5/5% Kraft 77 285 74 13 V-2 10% 5/10% Kraft 43 220 74 12 V-2 Lastly, although not directly related to the FR properties (but vastly important nonetheless), 15% 5/5% Organosolv 61 250 78 13 V-2 the authors10% 5/10% stated Organosolv that the incorporation46 of 5 and250 lignin had detrimental 65 effects on 13 the mechanicalV-2 properties. Tensile strength for PLA was1 measured No classification. at 70.3 MPa. For example, incorporation of PA and Kraft lignin separately and together decreased this value to 27.7, 16.9, and 35.8 MPa, respectively. ElongationWithin at the break, polymer as well industry, as impact vinyl resistance, polymers also make decreased. up a large bulk of the market, of which, polyacrylonitrileWithin the polymer(PAN) is industry,an important vinyl commodity polymers makein which up acrylonitrile a large bulk is of often the market,incorporated of which, as a polyacrylonitrilecomonomer with (PAN)monomers is an such important as styrene commodity [40]. Recently, in which Ren acrylonitrile and coworkers is often [41] incorporated utilized a sol-gel as a processcomonomer to prepare with monomers new FR organic–inorganic such as styrene [hybrid40]. Recently, coatings Ren that and can coworkers be applied [41 to] PAN utilized fabrics. a sol-gel Sol- processgel processes to prepare usually new involve FR organic–inorganic the hydrolysis of hybridreactive coatings inorganic that monomers can be applied in solution to PAN (“sol”) fabrics. that canSol-gel then processes condense usually to form involve crosslinked the hydrolysis networ ofk reactive(“gel”) that inorganic can be monomers molded inand solution shaped. (“sol”) The hydrolysisthat can then and condense subsequent to formreaction crosslinked of tetraethoxysilane network (“gel”) 13 is a that classic can example. be molded In the and current shaped. study, The thehydrolysis authors and applied subsequent this process reaction to of tetraethoxysilanethe preparation 13of isa asynergistic classic example. FR coating In the currentthrough study, the combinationthe authors applied of 5, urea this 14 process and 13 to as the a sol preparation which was of then a synergistic used to coat FR coating PAN fabrics through and the allowed combination to gel (Figureof 5, urea 10).14 and 13 as a sol which was then used to coat PAN fabrics and allowed to gel (Figure 10).

FigureFigure 10. 10. Sol-gelSol-gel process process to to prepare prepare 14–5 FR coating for polyacrylonitrile (PAN) fabrics.

The treated PANPAN fabrics fabrics were were then then analyzed analyzed to determineto determine their their thermal thermal stability stability using using TGA TGA and were and werecompared compared to untreated to untreated PAN fabric PAN (control). fabric The(contr controlol). The sample control exhibited sample a two-step exhibited decomposition: a two-step ◦ decomposition:ﬁrst, 32% weight first, loss between32% weight 276 and loss 366 betweenC resulting 276 fromand cyclization-dehydrogenation366 °C resulting from cyclization- reactions ◦ ◦ dehydrogenationand a maximum weightreactions loss and of 24%a maximum between 366 weight and 550lossC. of Carbonization 24% between occurred 366 and ca. 550 400 °C.C Carbonization occurred ca. 400 °C to leave behind a rather significant graphitic residue (ca. 44%). The

Polymers 2019, 11, 224 12 of 31 to leave behind a rather significant graphitic residue (ca. 44%). The PAN fabrics composed of Si, Si–5, and Si–5–14 sol-gels were also tested. The PAN treated with the Si sol-gel exhibited thermal behavior similar to that of the control but showed a higher thermal stability (due to the presence of Si–O bonds) and enhanced char formation composed of SiO2. The PAN treated with Si–5 sol-gel exhibited more complex behavior with 4 decomposition steps and exhibited early loss of water and lower onset temperature of cyclization (ca. 125 ◦C). This indicated that 5 promotes the cyclization of PAN. Maximum weight loss occurred at 324 ◦C. The lower decomposition temperature resulted in higher oxide char formations from the synergistic effect of the Si–O–Si and P–O–P bonds. This sample left a char residue of ca. 68% at 800 ◦C. The PAN fabric treated with Si–5–14 sol-gel exhibited a four-step decomposition as well. Early weight loss (between 62 and 110 ◦C) of water and ethanol, followed by the weight loss up to 255 ◦C (and a maximum weight loss at 171 ◦C) is ascribed to the decomposition of 14, which forms ammonia and can assist as a protective FR in the gas phase. Even though it left lower levels of char (ca. 56%), it does act as an effective FR. The LOI values for the control, Si, Si–PA, Si–5–14 sol-gels are 18, 25.1, 30.3, 34.1%, respectively. Unfortunately, it was found that these values decreased significantly after each wash cycle (for example, LOI for PAN fabric coated with Si–5÷14 sol-gel decreased to 23.4% after four washes). Layer-by-layer (LbL) assembly has been shown to be a relatively rapid and efficient process to prepare multifunctional coatings on a variety of different substrates [42]. The coatings are formed through repeatedly depositing alternating layers of anionic and cationic materials. In 2018, Ji et al. [43] showed that LbL assembly could be combined with a sol-gel process to prepare flexible (many sol-gels become brittle) FR polyester composites. The sol-gel process was utilized to prepare a flexible polysiloxane (SSP) “sol” from methyltriethoxysilane (MTES). The LbL assembly was carried out by immersion of polyester fabric in the sol solution of SSP for 30 s, followed by washing with water. After drying, the sample was immersed in a PA solution (PAS) (for 0 to 20 min) and subsequently washed with water. Then it was immersed in SSP, washed, and dried. This process was carried out three times to yield a flexible polysiloxane coated polyester composite. The TGA showed that the amount of PA (controlled by varying the time the sample was immersed in PAS) could have a drastic effect on the thermal stability of the polymer samples. Pure polyester fabric showed a simple two-step decomposition to leave behind no residue. Soaking the sample in PAS for 5, 10, 15, and 20 min showed an increase in the residue after combustion to 17.3, 18.4, 18.5, and 21.5%, respectively. The Tmax was also affected and could be decreased to 439.8 ◦C (sample soaked for 20 min) from 441.3 ◦C (control). To gain more insight into the FR properties, LOI measurements and vertical flame tests were performed and the data is presented in Table6. Upon ignition, the control sample burned completely and rapidly (burn rate of 1.19 mm/s) with an LOI of 21.6%. The authors noted that in all cases the spread of the flame was not stopped, but all samples burned to the top. However, the inclusion of more PA (as indicated by the soak time) did have some dramatic effects on the LOI values (up to 31.4%) and burning rates (down to 0.32 s). Further, the melt-dripping (a potentially devastating effect caused by house fires) could be reduced from 44% to 16.4%. Advantageously, the authors note that this LbL, sol-gel combination produced much more durable FR fabrics than treating a sample with PA alone. They showed that the amount of PA remained almost constant in most samples, even after 45 wash cycles. This was illustrated by LOI measurements after each wash cycle and it was shown that the worst-case scenario resulted in a decrease of only ca. 3%. Another industrially-important vinyl polymer is poly(vinylchloride), or PVC. This material has become incredibly important and widely used for a variety of applications (for example, piping and flexible laboratory tubing) [44]. However, because of the presence of flammable plasticizers, flexible PVC can catch fire and burn rapidly [45]. Polymers 2019, 11, 224 13 of 31

Table 6. Vertical burn test data for FR polyester sol-gel composites prepared using layer-by-layer (LbL) assembly.

Sample After Flame (s) Char Length (mm) LOI (%) Burning Rate (mm/s) Melt-Dripping (%) Control 44.0 26.9 21.6 1.19 44.0 5 min soak 30.2 26.4 30.3 0.35 30.2 10 min soak 25.0 16.1 30.5 0.34 25.0 15 min soak 15.9 13.5 30.8 0.33 15.9 20 min soak 16.4 12.3 31.4 0.32 16.4

Recently, Qu’s [46] lab showed that 5-metal salts can be used as effective FR additives to PVC. The flexible PVC composites were prepared by mixing PVC with 40% dioctylphthalate (DOP, as plasticizer), 4% organic tin stabilizer, 0.5% calcium stearate, 0.5% stearic acid, and 15% PA–metal complex (either Zn, Cu, Al, or Sn). The melt was mixed, compressed into a sheet, and cooled. The FR properties of each sample were measured by cone calorimetry. The data is summarized in Table7. The TTI showed little variation between PVC (control) and composites. On the other hand, LOI values could be increased from 24.9% (for the control) up to 30.3% for PVC-5-Sn composite. The inclusion of 5-M complexes also had a significant effect on the protective char formation: control PVC burned to near completion while the 5–Cu composite left ca. 19% residue. Furthermore, it was observed that PHRR could be decreased by almost 150 kW/m2 (from 329 to 181.77 kW/m2 for the control and 5–Cu composite, respectively). Considering that most injuries and deaths attributable to fires is actually smoke inhalation, the total smoke production (TSP) of a FR material is very important and cannot be overlooked. As can be seen from Table7, the TSP could be decreased from 42.20 to as low as 15.77 m 2 for the control and Cu composite, respectively.

Table 7. Cone calorimetry data for poly(vinylchloride) (PVC)–5-metal composites.

Sample TTI (s) Char Residue (%) LOI (%) PHRR (kW/m2) TSP (m2) Control 13 1.68 24.9 329.67 42.20 Sn 14 14.68 30.3 213.75 19.51 Zn 14 17.07 29.7 225.25 23.16 Cu 10 18.93 29.3 181.77 15.77 Al 15 17.50 27.3 245.34 25.45

2.3. Isosorbide Isosorbide 15 is bicyclic diol that is can be obtained directly from D-sorbitol, which is ultimately derived from starch. This compound has, in recent years, attracted attention as a bio-based monomer for several polymeric materials [47]. Because of the presence of two hydroxyl groups, 15 can be instilled with various functionality. Recently, it has been shown that this can be taken advantage of for the preparation of new, bio-based phosphorus FR. Boday et al. [48] showed that a polyphosphonate 17 could be prepared through a stepwise, condensation polymerization reaction between 15 and phenylphosphonic dichloride 16 in the presence of N,N-dimethylaminopyridine (DMAP), according to Figure 11[ 48]. Upon preparation, 17 (Mw = 1020 g/mol) was blended with PLA (Mw = 4032 g/mol) by solvent casting in chloroform, followed by removal of solvent under elevated temperature and reduced pressure. Cone calorimetry was performed to obtain thermal stability information for blends of varying 17 content. Under a heat ﬂux of 25 kW/m, little difference was observed between PHRR values of control PLA and PLA-17 blends (PLA = 786, 5%, 10%, and 15% 17 = 796, 744, and 788 kW/m2). The THR was affected by the inclusion of 17. Control PLA exhibited a value of 105 MJ/m2 while the 10% 17 blend showed a value of 82 MJ/m2. Further, the TTI decreased upon incorporation of 17. To test the FR properties under more realistic conditions, the authors turned to UL-94 vertical ﬂame tests. The PLA burned completely and could not be rated. PLA-17 blends of 5, 10, and 15% 17 exhibited low extinguishing time (1.40, 0.80, and 0.10 s, respectively). The 15% 17 blend showed a V0 rating Polymers 2018, 10, x FOR PEER REVIEW 13 of 30 flexible laboratory tubing) [44]. However, because of the presence of flammable plasticizers, flexible PVC can catch fire and burn rapidly [45]. Recently, Qu’s [46] lab showed that 5-metal salts can be used as effective FR additives to PVC. The flexible PVC composites were prepared by mixing PVC with 40% dioctylphthalate (DOP, as plasticizer), 4% organic tin stabilizer, 0.5% calcium stearate, 0.5% stearic acid, and 15% PA–metal complex (either Zn, Cu, Al, or Sn). The melt was mixed, compressed into a sheet, and cooled. The FR properties of each sample were measured by cone calorimetry. The data is summarized in Table 7. The TTI showed little variation between PVC (control) and composites. On the other hand, LOI values could be increased from 24.9% (for the control) up to 30.3% for PVC-5-Sn composite. The inclusion of 5-M complexes also had a significant effect on the protective char formation: control PVC burned to near completion while the 5–Cu composite left ca. 19% residue. Furthermore, it was observed that PHRR could be decreased by almost 150 kW/m2 (from 329 to 181.77 kW/m2 for the control and 5–Cu composite, respectively). Considering that most injuries and deaths attributable to fires is actually smoke inhalation, the total smoke production (TSP) of a FR material is very important and cannot be overlooked. As can be seen from Table 7, the TSP could be decreased from 42.20 to as low as 15.77 m2 for the control and Cu composite, respectively.

Table 7. Cone calorimetry data for poly(vinylchloride) (PVC)–5-metal composites.

Polymers 2018, 10, x FOR PEER REVIEW 14 of 30

The PLA burned completely and could not be rated. PLA-17 blends of 5, 10, and 15% 17 exhibited low extinguishing time (1.40, 0.80, and 0.10 s, respectively). The 15% 17 blend showed a V0 rating while the other blends were rated at V2. Unfortunately, the authors note that the increased content of 17 had deleterious effects on the mechanical properties. It was shown that the tensile strength and strain-at-break of the blends decreasedFigure 11. 3-fold.Synthesis of polyphosphonate 17.17. Recently, Howell and Daniel [49] were able to synthesize a variety of phosphate esters (18–21) fromConeRecently, isosorbide calorimetry Howell that could and was Daniel be performed used [49 as] wereFR to additive obtain able to thermal synthesizes for epoxy stability a resins variety information (Figure of phosphate 12). forThe estersblends phosphate (18 of– 21varying )esters from 17couldisosorbide content. easily that Under be couldprepared a heat be used directlyflux as of FR 25 from additives kW/m, isosorbide little for epoxydi throughfference resins reactionwas (Figure observed with12). Thethe between phosphaterespective PHRR estersphosphoryl values could of controlchloride.easily be PLA prepared and PLA- directly17 blends from isosorbide(PLA = 786, through 5%, 10%, reaction and 15% with 17 the = respective796, 744, and phosphoryl 788 kW/m chloride.2). The THR was affected by the inclusion of 17. Control PLA exhibited a value of 105 MJ/m2 while the 10% 17 blend showed a value of 82 MJ/m2. Further, the TTI decreased upon incorporation of 17. To test the FR properties under more realistic conditions, the authors turned to UL-94 vertical flame tests.

Figure 12. IsosorbideIsosorbide phosphate esters used as FR additives.

Upon their synthesis, the authors studied the th thermalermal stability of each derivative using TGA, which did not seem to reveal any clear trend betweenbetween structure and thermal stability. However, the ◦ authors note that each compound revealed interesting isothermal degradation at 200 °C,C, in which phosphonate 18 seemed to be quite stable, losing less than than 10% 10% mass mass after after 18 18 hours hours of of heat heat exposure. exposure. Phosphonite 20 also exhibited some stability, stability, decomposing by by roughly roughly 40% 40% after after 15 15 h h of of exposure. exposure. Phosphates 1919and and21 underwent21 underwent relatively relatively rapid decompositionrapid decomposition to lose the to corresponding lose the corresponding phosphorus phosphorusacid. In fact, acid. all ofIn thefact, studied all of the compounds studied compound lost phosphoruss lost phosphorus acid initially, acid initially, just to different just to different extents. extents.Because Because of this, theof authorsthis, the hypothesizedauthors hypothesized that such that compounds such compounds may be in may an effective be in an condensed effective condensedFR phase for FR epoxy phase resins.for epoxy To testresins. this, To each test compoundthis, each compound was dissolved was indissolved bisphenol in Abisphenol diglycidyl A ◦ diglycidylether 22 epoxy ether and22 epoxy then and cured then with cured 2-ethyl-4-methylimidazole with 2-ethyl-4-methylimidazole23 at 95 23C at to95 provide°C to provide resins resins with 1with or 2%1 or phosphorus2% phosphorus content content (Figure (Figure 13). 13). The The thermal thermal properties properties of of each each resinresin werewere then tested. tested. The thermal stability of the resin was not dramatically affected by the presence of the additives with a few exceptions: the the addition addition of of 1919 and 21 at loadings of 2 and 3.4% phosphorus provided onset ◦ temperatures of 329 and 319 °CC in nitrogen, respectivelyrespectively (resin with no additive exhibited an onset ◦ temperature ofof 390390 °C).C). TheThe char char formation formation could could be be increased increased from from 8% 8% (with (with no additive)no additive) to 26% to 26% (2% (2%20). The20). The LOI LOI value value for resin for resin containing containing no additive no additive was 18.8%.was 18.8%. This couldThis could be increased be increased to 31.6% to 31.6% upon uponthe addition the addition of 1% of20 .1% Fortunately, 20. Fortunately, resins containingresins containing 1% 18 and1% 1820 andreached 20 reached V1 rating V1 whilerating 2% while18 and 2% 1918 reachedand 19 reached V0 in UL-94 V0 invertical UL-94 vertical burn tests. burn tests.

Figure 13. Preparation of FR epoxy resins from 22 and 23 containing 18, 19, 20, or 21.

2.4. Diphenolic Acid

Polymers 2018, 10, x FOR PEER REVIEW 14 of 30

Figure 12. Isosorbide phosphate esters used as FR additives.

Upon their synthesis, the authors studied the thermal stability of each derivative using TGA, which did not seem to reveal any clear trend between structure and thermal stability. However, the authors note that each compound revealed interesting isothermal degradation at 200 °C, in which phosphonate 18 seemed to be quite stable, losing less than 10% mass after 18 hours of heat exposure. Phosphonite 20 also exhibited some stability, decomposing by roughly 40% after 15 h of exposure. Phosphates 19 and 21 underwent relatively rapid decomposition to lose the corresponding phosphorus acid. In fact, all of the studied compounds lost phosphorus acid initially, just to different extents. Because of this, the authors hypothesized that such compounds may be in an effective condensed FR phase for epoxy resins. To test this, each compound was dissolved in bisphenol A diglycidyl ether 22 epoxy and then cured with 2-ethyl-4-methylimidazole 23 at 95 °C to provide resins with 1 or 2% phosphorus content (Figure 13). The thermal properties of each resin were then tested. The thermal stability of the resin was not dramatically affected by the presence of the additives with a few exceptions: the addition of 19 and 21 at loadings of 2 and 3.4% phosphorus provided onset temperatures of 329 and 319 °C in nitrogen, respectively (resin with no additive exhibited an onset temperature of 390 °C). The char formation could be increased from 8% (with no additive) to 26% (2% 20). The LOI value for resin containing no additive was 18.8%. This could be increased to 31.6%

Polymersupon the2019 addition, 11, 224 of 1% 20. Fortunately, resins containing 1% 18 and 20 reached V1 rating while15 of 2% 31 18 and 19 reached V0 in UL-94 vertical burn tests.

Figure 13. Preparation of FR epoxy resins from 22 and 23 containing 18, 19,, 20,, oror 21.. 2.4.Polymers Diphenolic 2018, 10, x Acid FOR PEER REVIEW 15 of 30 2.4. Diphenolic Acid Diphenolic acid 24 is a bio-mass bio-mass derived derived diol that that can can be be prepared prepared from reaction between phenol and levulinic acid, acid, which which can can be be produced produced from from the the degradation degradation of of cellulose cellulose [50]. [50 Diphenolic]. Diphenolic acid acid 24 24is anis aninteresting interesting structure structure because because of the of the functiona functionalitylity that that can can potentially potentially be beincorporated incorporated into into its itsframework framework through through the thehydroxy hydroxy and andcarboxylic carboxylic acids acids groups. groups. Fang Fangand coworkers and coworkers [51] have [51] taken have takenadvantage advantage of this ofand this prepared and prepared a novel a novelbio-based bio-based polyphosphonante polyphosphonante 27 that27 canthat be can used be as used an aseffective an effective FR additive FR additive for PLA. for PLA.Polymer Polymer 28 could28 could easilyeasily be prepared be prepared in two in steps, two steps, beginning beginning with withreaction reaction between between 24 and24 and 1-phosphabicyclo[2.2.2]octane-4-methanol 1-phosphabicyclo[2.2.2]octane-4-methanol 25 25(to(to form form 2626) )followed followed by condensation polymerizationpolymerization with with phenylphosphonic phenylphosphonic acid acid dichloride dichloride27 (Figure 27 (Figure 14). Blends 14). ofBlends28/PLA of ◦ were28/PLA prepared were prepared in a melt in mixer a melt at mixer 170 C at at 170 60 °C rpm at for60 rpm 8 min for with 8 min 2, 4,with and 2, 6 4, wt and % 286 wtto % provide 28 to provide blends PLAblends 2, PLA 2, 4, andPLA PLA4, and 6, respectively.PLA 6, respectively.

Figure 14. Preparation of 2828..

The UL-94UL-94 vertical vertical burn burn tests tests were were performed performed and LOIand valuesLOI values were measured.were measured. Pure PLA Pure (control) PLA exhibited(control) exhibited an LOI of an 20.0 LOI while of 20.0 PLA while 2, PLA PLA 4, and2, PLA PLA 4, 6 and were PLA 28.8, 6 33.7,were and 28.8, 35.4, 33.7, respectively. and 35.4, Therespectively. UL-94 vertical The UL-94 burn testsvertical of PLAburn (control) tests of PL failedA (control) while PLA failed 2, PLAwhile 4, PLA and PLA2, PLA 6 were 4, and classified PLA 6 aswere V2, classified V0, and as V0, V2, respectively. V0, and V0, Conerespectively. calorimetry Cone was calorimetry used to studywas used the combustionto study the behaviorcombustion of 2 allbehavior samples. of all The samples. data is The summarized data is summarized in Table8 in. PureTable PLA 8. Pure exhibited PLA exhibited a PHRR a ofPHRR 418 kW/mof 418 kW/mand2 2 incorporationand incorporation of up of to up 6% to28 6%only 28 slightly only slightly decreased decreased this value this (388 value kW/m (388 ).kW/m Furthermore,2). Furthermore, TTI showed TTI almostshowed insignificant almost insignificant differences differences between between the samples. the samples. The authors The noteauthors this note discrepancy this discrepancy between UL-94between and UL-94 LOI testsand LOI and tests calorimetry and calorimetry and explained and explained that these that inconsistencies these inconsistencies arise from arise the differentfrom the firedifferent scenarios fire scenarios recreated recreated with each with method. each method.

Table 8. Cone calorimetry data for PLA–28 blends. Table 8. Cone calorimetry data for PLA–28 blends. 2 2 SampleSample TTI TTI (s) (s) PHRRPHRR (kW/m (kW/m2) ) ResidualResidual mass Mass (wt %) (wt %) THRTHR (MJ/m (MJ/m2) ) ControlControl 68 68 418 418 1.6 1.6 70 70 PLAPLA 22 68 68 394 394 2.5 2.5 69 69 PLAPLA 44 78 78 396 396 3.1 3.1 67 67 PLAPLA 66 74 74 388 388 3.5 3.5 66 66

Intrigued by the positive burn tests, the authors were prompted to test the FR mechanism at play. To do this, TGA was employed. Under N2 atmosphere, pure 28 showed a Tonset of ca. 237 °C and left a char residue of 41.40 wt % at 600 °C. This sample exhibited two Tmax values (ca. 252 and 388 °C). Pure PLA exhibited a Tonset of ca. 344 °C and a Tmax of ca. 379 °C with much lower residue (1.72 wt %) than 28. The Tonset of the PLA blends decreased with increasing 28, as is common with phosphorus FR but showed no significant differences in Tmax values or residual char residue. The authors noted that these results, combined with the inconsistencies between experimental and theoretical residue, point toward a gas-phase FR mechanism. Additionally, TG-FTIR was utilized to study the evolved gases during thermal decomposition for 28. At Tonset (237 °C), a carbonyl stretch at 1775 cm−1 was observed, which indicates that the ester group was broken. At the first Tmax (252 °C), stretches associated with water (3660 cm−1), hydrocarbons (2970 cm−1), alkenes (1605 cm−1), ethers/alcohols (1101 cm−1), and P=O (1257 cm−1) were observed. Furthermore, at 388 °C signals associated with alkenyl C–H stretches were observed, indicating products evolved during char formation. The authors conclude that this serves

Polymers 2019, 11, 224 16 of 31

Intrigued by the positive burn tests, the authors were prompted to test the FR mechanism at play. ◦ To do this, TGA was employed. Under N2 atmosphere, pure 28 showed a Tonset of ca. 237 C and left a ◦ ◦ char residue of 41.40 wt % at 600 C. This sample exhibited two Tmax values (ca. 252 and 388 C). Pure ◦ ◦ PLA exhibited a Tonset of ca. 344 C and a Tmax of ca. 379 C with much lower residue (1.72 wt %) than 28. The Tonset of the PLA blends decreased with increasing 28, as is common with phosphorus FR but showed no significant differences in Tmax values or residual char residue. The authors noted that these results, combined with the inconsistencies between experimental and theoretical residue, point toward a gas-phase FR mechanism. Additionally, TG-FTIR was utilized to study the evolved gases during ◦ −1 thermal decomposition for 28. At Tonset (237 C), a carbonyl stretch at 1775 cm was observed, which ◦ indicates that the ester group was broken. At the first Tmax (252 C), stretches associated with water (3660 cm−1), hydrocarbons (2970 cm−1), alkenes (1605 cm−1), ethers/alcohols (1101 cm−1), and P=O (1257 cm−1) were observed. Furthermore, at 388 ◦C signals associated with alkenyl C–H stretches were observed, indicating products evolved during char formation. The authors conclude that this serves as evidence that phosphorus-containing radical scavengers exist in the evolved gases. The PLA, on the −1 −1 −1 other hand, showed water (3580 cm ), hydrocarbons (3002–2850, 1400–1200 cm ), CO2 (2341 cm ), CO (2182 cm−1), and carbonyls (1761 cm−1). The PLA blends exhibited TG-FTIR spectra that were very similar to that of PLA, with the exception of a P=O stretch (1258 cm−1) for PLA 6. A couple of years later, this same group described the application of 28 as an FR additive to PLA along with modified graphene oxide (M-GO) [52]. They chose this material because it has been shown to improve mechanical properties as well as impart some FR behavior because of its layered and graphitized structure. Because synergistic intumescent materials can be made with P and N-containing FR, graphene oxide (GO) was modified with 7 in the presence of 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC). Composites were prepared through solution mixing of 80:20 ratio of PLA and M-GO in chloroform. The samples were dried and subjected to melt mixing with 28 at 160 ◦C to obtain composites containing 3 wt % of 28/M-GO total with varying ratios of 28:M-GO (3.0:0.0, 2.7:0.3, 2.4:0.6, and 2.1:0.9). The authors first tested the flammability of each sample. The results are summarized in Table9. The LOI values were measured and compared to the control PLA. These values could be improved upon the addition of FR 28 and M-GO up to 35.6%. However, maintaining the amount of 28:M-G at 3%, the LOI decreased as the amount of 28 decreased. Compared to PLA, all samples reached V0 or V2 ratings in the UL-94 burn test. The authors note that both samples containing no M-GO exhibited heavy dripping during the test. The incorporation of M-GO inhibited this dripping. The authors note that this was due to the increase in viscosity within the polymer matrix. Further, M-GO may slow the mass transfer in the condensed phase, leading to less dripping. Considering that 28 serves, mainly, as a gas-phase FR, the presence of M-GO can decrease the amount and speed of volatilization of decomposed gas product evolution. Cone calorimetry showed that each composite exhibited improved FR properties over PLA alone. The PHRR values could be decreased from 393 to 373 kW/m2 for PLA and PLA-2.1:0.9, respectively.

Table 9. Flammability of PLA composites of 28.

Sample (28:M-GO) LOI (%) UL-94 PLA 20 NR PLA-3:0 33.6 V0 PLA-2.7:0.3 35.3 V0 PLA-2.4:0.6 36.0 V0 PLA-2.1:0.9 32.7 V2

2.5. Deoxyribonucleic Acid (DNA) As mentioned earlier, to be considered an IFR, materials must contain an acid source, charring agent (carbon source), and a blowing agent. Some of the most widely used and successful IFR materials are based on phosphorus and nitrogen-containing compounds. Some examples of these have already Polymers 2019, 11, 224 17 of 31 been discussed and may rely on a synergistic relationship between a P-compound (acid source, such as 8) with an N-compound (blowing agent, such as 10). However, preparing an IFR additive or polymer that contains both may be attractive. Even better would be using a naturally-occurring, bio-based compound that contains the acid source and blowing agent. Considering that its structure is based on nitrogenous base pairs, sugars and phosphate groups, deoxyribonucleic acid (DNA) contains all the elements required to act as an IFR. In fact, DNA has been used to instill FR properties to polymeric materials. Malucelli et al. [53,54] reported the first examples describing the utilization of DNA (extracted from Herring) as an FR coating for cotton fibers. That same laboratory then reported the first examples of DNA used in conjunction with synthetic polymers. This group described the preparation of FR ethylene vinyl acetate (EVA) composites utilizing DNA as an IFR additive [55]. The blends (containing 10, 15, and 20 wt % DNA) were prepared by melt-blending in a mixer at 120 ◦C and 60 rpm for 3 min. Once prepared, the authors analyzed the thermal stability of each sample using TGA. Under nitrogen, control EVA exhibited two degradation steps of maximum weight loss at 350 and 472 ◦C. Inclusion of 15% DNA reduced the onset temperature ◦ and exhibited a new Tmax at 240 C from the decomposition of DNA. Importantly, it was shown that residue char formation at 600 ◦C increased with increasing DNA content: 10, 15, and 20 wt % DNA showed 6.0, 9.0, and 11.0% residue, respectively. Control EVA left behind less than 1.0% residue under nitrogen. The authors were also interested in investigating any synergistic effects of DNA with other carbon sources. They chose to study two other bio-based sources, α-cellulose (CEL) and β-cyclodextrin (βCD). The TGA showed that sample containing 10 wt % DNA and 10 wt % CEL (as well as samples containing 10 wt % DNA and 10 wt % βCD) exhibited behavior very similar to that of the sample containing 20% DNA. The FR properties were studied using cone calorimetry. Control EVA combustion was rapid with PHRR of 1588 kW/m2. The incorporation of DNA had a significant impact on this combustion. Although the TTI values decreased (from 65 to 28 s for EVA and EVA-10% DNA, respectively), the blends with DNA burned much more slowly and promoted the formation of protective char. Furthermore, the PHRR could significantly be reduced to 973 and 963 kW/m2 for the blends containing 10 and 20 wt % DNA. Although, the THR and TSR were almost the same as EVA alone. Blends of EVA, DNA, and CEL exhibited similar characteristics. Even LOI values for all blends were similar (between 20–22%). However, the sample containing 10 wt % DNA and 10 wt % CEL did show reduced CO and CO2 formation. To improve on these results, the same laboratory later studied the differences in thermal and FR properties of impregnating versus simply coating the surface of EVA copolymers with DNA [56]. To coat the surface of EVA with DNA, DNA was deposited on EVA square plates followed by compression using a hot molding press at 120 ◦C and 5 MPa. Cone calorimetry was used to assess the resistance to heat flux of 35 kW/m2. As described earlier, the incorporation of DNA through melt blending reduced the TTI of EVA (from 62 to 28 s) and also showed a decrease of ca. 28% in the PHRR. Alternatively, when DNA is simply placed on the EVA surface, it takes up to 5 min for the sample to ignite. This corresponds to a 380% increase in TTI. This sample burned slowly with a PHRR of 348 kW/m2. The THR also decreased from 106 to 67 MJ/m2 for EVA–DNA melt blend and EVA coated with DNA, respectively. Furthermore, TSR was also reduced (from 1637 to 1520 m2/m2). These findings suggest that the protective char layer on the surface was able to protect the copolymer from combustion more efficiently than that of the sample prepared from melt-blending. The authors also studied each sample’s resistance to combustion using a burning-through test. Each sample was placed in a vertical position and flame from a butane lighter was applied three consecutive times. Each time the temperature of the sample was measured; it was found that EVA alone reached a maximum of 346 ◦C upon the third flame application while surface-modified EVA never surpassed 60 ◦C. Furthermore, the sample prepared using melt blending reached 188 ◦C. Alongi [57] argued that this surface protection method could be a viable alternative to the bulk inclusion of FR additives to polymers and that any polymeric surface could, in theory, be coated without Polymers 2019, 11, 224 18 of 31 affecting the bulk polymer’s properties. To test this, Alongi’s laboratory applied DNA coatings to a variety of polymeric materials and measured their thermal stability as well as FR properties. Samples of polypropylene (PP), acrylonitrile-butadiene-stryene (ABS) resin, poly(ethyleneterephthalate) (PET), and polyamide 6 (PA6) were coated with DNA using hot compression molding press at 120 ◦C for 1.5 min. Each sample had a thickness of 3 mm and contained 10 wt % DNA. Cone calorimetry was employed to test the behavior of each material at heat flux of 35 and 50 kW/m2. In every case it was shown that PHRR could be significantly reduced while TTI values increased, similar to what was observed for EVA (Table 10). For example, PP shows a ca. 50% decrease in PHRR under either 35 or 50 kW/m2 heat flux.

Table 10. Flammability of polymers surfaced-treated with DNA.

Heat Flux PA Property PP PP–DNA PET PET–DNA ABS ABS–DNA PA 6 (kW/m2) 6-DNA PHRR 1300 629 973 563 1180 512 1036 562 35 (kW/m2) TTI (s) 15 156 82 274 26 344 19 330 PHRR 1800 900 892 525 1479 433 1488 810 50 (kW/m2)

Polymers 2018, 10, x FORTTI PEER (s) REVIEW 32 58 27 246 12 80 48 166 18 of 30

Pokorski and coworkers [58] [58] were able to prepareprepare FR low density polyethylene (LDPE, one of the most widely used polymers worldwide)–DNA worldwide)–DNA composites composites through through melt-blending. melt-blending. Additionally, Additionally, the authorsauthors compared compared all all thermal thermal data data to LDPE to LD compositesPE composites of melamine-polyphosphate of melamine-polyphosphate29, a common 29, a commonindustrial industrial FR, the structure FR, the structure of which isof shownwhich inis shown Figure 15in .Figure 15.

Figure 15. Chemical structure of 29.

Thermal analysis was performed using TGA. Under nitrogen, LDPE undergoes complete ◦ decomposition atat ca.ca. 433433 °CC toto leaveleave no no char char residue. residue. Samples Samples of of LDPE–DNA LDPE–DNA and and LDPE– LDPE–29 29blends blends (at (atloadings loadings of 5,of 10,5, 10, 20, 20, 30, 30, and and 40 40 wt wt % % DNA DNA or or29 29) exhibited) exhibited two two major major weight weight lossloss events:events: for DNA ◦ ◦ blends the first ﬁrst occurred around 200 °CC and for 29 around 350 °CC (due (due to to decomposition decomposition of of additive) additive) ◦ while the second event occurred around 430 °CC for both (due to LDPE decomposition). Char residue yields for pure DNA and 29 were 41.7 and 31%, respectively. TheThe blends exhibited char yields from 1.3 and 2% (5% DNA or 29) upup toto 11.411.4 andand 12.1%12.1% (40%(40% DNADNA oror 29).). Interestingly, mico-combustionmico-combustion calorimetry (MCC) showed that neither DNA nor 29 caused any dramatic change in PHR, all values ◦ hovered around 500 °C.C. The PHHR (W/g) of of LDPE LDPE wa wass 1123.4 1123.4 and and could could be be decreased decreased with with increasing increasing DNA or 29 content (695.5 and 701.6701.6 W/gW/g for for 40 40 wt wt % % DNA and 29, respectively). The UL-94 vertical burn tests revealed FR properties of thethe polymerpolymer blends.blends. It showed that the LDPE–DNA blends exhibited significantly signiﬁcantly reduced burn distance comparedcompared toto 2929 blends.blends. This self-extinguishing is probably due to the early DNA decomposition. Addition Additionally,ally, it acts as a carbon source to facilitate char formation. In 2017, Li and coworkers [59] [59] described the preparation and use of a synergistic FR additive based on nitrogenous bases (the major components of DNA) and IFR system (composed of APP and pentaerythritol 9). The The PP PP composites composites were were prepared prepared by by melt melt blending blending PP, PP, nitrogenous nitrogenous base, base, and and IFR. IFR. The samples were then analyzed by UL-94 vertical burn testing. Pure PP exhibited, unexpectedly, rapid ignition and heavy dripping (no classification) with an LOI value of 17.5%. The PP composite with IFR (18 wt %) increased the LOI to 24.5% but also could not be classified. At 25 wt % a V0 rating could be obtained with an LOI value of 28.3%. However, these exhibited some melt dripping. Incorporation of nitrogenous bases could, in some instances, improve the FR properties. For instance, a PP composite of 17 wt % IFR and 1 wt % cytosine (C) showed no improvement over the LOI value (27.9%) and had the same V0 rating, but melt dripping was not observable. Combustion behavior was further studied using cone calorimetry. Pure PP exhibited a PHRR value of 888 kW/m2 while the incorporation of 18 wt % IFR reduced this value greater than two-fold. This could further be improved to 324 and 293 kW/m2 through the inclusion of 1 wt % guanine (G) and uracil (U), respectively.

2.6. Lignin As noted earlier, lignin is a structurally-complex, highly crosslinked, naturally occurring phenolic polymer found in wood and bark. The exact chemical structure of lignin can vary dramatically based on the species (and part) of the plant from which it is extracted. Lignin is among the most abundant polymers in plant materials and is an abundant byproduct of paper production (approximately 50,000,000 tons of this material are produced each year). But, its variation in structure and amorphous composition has historically limited its use in valuable industrial processes (at the end of the 20th century, only ca. 1% of lignin produced was used in this manner) [60]. However, the

Polymers 2019, 11, 224 19 of 31

The samples were then analyzed by UL-94 vertical burn testing. Pure PP exhibited, unexpectedly, rapid ignition and heavy dripping (no classiﬁcation) with an LOI value of 17.5%. The PP composite with IFR (18 wt %) increased the LOI to 24.5% but also could not be classiﬁed. At 25 wt % a V0 rating could be obtained with an LOI value of 28.3%. However, these exhibited some melt dripping. Incorporation of nitrogenous bases could, in some instances, improve the FR properties. For instance, a PP composite of 17 wt % IFR and 1 wt % cytosine (C) showed no improvement over the LOI value (27.9%) and had the same V0 rating, but melt dripping was not observable. Combustion behavior was further studied using cone calorimetry. Pure PP exhibited a PHRR value of 888 kW/m2 while the incorporation of 18 wt % IFR reduced this value greater than two-fold. This could further be improved to 324 and 293 kW/m2 through the inclusion of 1 wt % guanine (G) and uracil (U), respectively.

Polymerscomposition 2018, 10 has, x FOR historically PEER REVIEW limited its use in valuable industrial processes (at the end of the19 20thof 30 century, only ca. 1% of lignin produced was used in this manner) [60]. However, the highly aromatic highlynature aromatic of lignin structuresnature of lignin allows structures for high levelsallows of for char high formation levels of uponchar formation combustion. upon For combustion. this reason, Forthe this use reason, of lignin the as use a bio-basedof lignin as FR a bio-based and antioxidative FR and antioxidative additive for additive synthetic for polymerssynthetic polymers has been hasinvestigated been investigated since the since early the 2000s. early In2000s. addition In addition to its charto its formation, char formation, its antioxidative its antioxidative activity activity can canbe attributed be attributed to its to high its high reactivity reactivity towards towards free free radicals. radicals. Song and coworkers [61] [61] described the preparatio preparationn of a chemically-modified chemically-modiﬁed lignin derivative. Commercially-available lignin lignin (isolated (isolated from from wheat straw) was grafted with nitrogen and phosphorus-containing compoundscompounds to to improve improve the the intumescent intumescent properties. properties. This This was was achieved achieved by ﬁrst by firsttreating treating lignin lignin with with formaldehyde formaldehyde in the in presencethe presence of NaOH of NaOH (aq). (aq). This This hydroxymethylated hydroxymethylated lignin lignin30 was30 was then then subjected subjected to reactionto reaction with with the the31 to31 formto form32 (Figure32 (Figure 16). 16). This This material material was was characterized characterized by byIR spectroscopyIR spectroscopy and and XPS. XPS.

Figure 16. Preparation of 32.

Before this material was used as an FR additive,additive, its thermal stability was firstfirst tested. The TGA ◦ showed that that unmodified unmodified lignin lignin (control) (control) started started to to decompose decompose at atca. ca. 250 250 °C. AC. maximum A maximum weight weight loss ◦ ◦ (lossTmax ()T wasmax) exhibited was exhibited at 405 at 405°C andCand a char a char residue residue of ofca. ca. 41% 41% was was left left behind behind at at 600 600 °C.C. The The TGA revealed 30 toto havehave slightlyslightly higherhigher stabilitystability withwith aa charchar residueresidue ofof ca.ca. 51%.51%. Modified Modified lignin 32 ◦ exhibited anan initial initial degradation degradation (5% (5% weight weight loss) loss) at 240 at C,240 with °C, awith more a complex more complex degradation degradation pathway, pathway,leaving behind leaving almost behind 62% almost char 62% residue. char residue. The ability The ofability this of material this material to act to as act an as FR an additive FR additive was wasthen then tested. tested. Modified Modified lignin lignin32 32was was incorporated incorporated into into poly(propylene- poly(propylene-coco-ethylene)-ethylene) (PPE)(PPE) samples using melt blending. The The thermal thermal stabilities stabilities of of th theseese composites composites were were then then analyzed analyzed using using TGA. TGA. ◦ Virgin PPE (control) exhibited initial initial decomposition decomposition ( (TTi)i )and and TTmaxmax valuesvalues of of324 324 and and 418 418 °C andC and left leftno ◦ charno char residue. residue. Incorporation Incorporation of 20 of 20and and 30 30wt wt % %of of3232 toto PPE PPE lead lead to to increases increases in in TTi i(325(325 and and 361 361 °C,C, ◦ respectively) andandT Tmaxmax (479(479 andand 483483 C,°C, respectively). respectively). These These higher higher values values are are related related to the to higherthe higher char charformations. formations. The sampleThe sample containing containing 20 wt %2032 wtleft % behind 32 left ca. behind 12% char. ca. This12% valuechar. wasThis comparable value was comparableto polymer blendsto polymer with blends 30 wt % with pure 30 lignin, wt % showingpure lignin, that showing the incorporation that the incorporation of P and N can of P increase and N can increase the thermal stability and char formation. Cone calorimetry was used to measure the FR properties of these materials. It was shown that virgin PPE burns rapidly with a time to ignition (tign) of 49 s, PHRR of 1350 kW/m2, and THR of 87.3 MJ. Incorporation of 20 wt % 32 exhibited tign of 38 s, PHRR of 380 kW/m2, and THR of 74.2 MJ while 30 wt % PN-lignin incorporation values were 39, 360, and 69.7. Both cases revealed dramatic decreases in PHRR (ca. 70% decrease) and THR. Hu et al. [62] used a similar strategy for the preparation of synergistic FR systems based on chemically-modified lignin PU foams and ammonium polyphosphate 8. Wheat straw lignin was first converted to 30, which was then treated with phosphoryl chloride, followed by ethylene glycol to form modified lignin 33 (Figure 17). The PU foams were prepared by mixing 33 and PFAPP (microencapsulated 8, prepared from reaction with phenol, formaldehyde and 8) with methylene diphenyl diisocyante (MDI) and polyol in various amounts. The thermal behavior of the resulting materials was then investigated.

Polymers 2018, 10, x FOR PEER REVIEW 19 of 30 highly aromatic nature of lignin structures allows for high levels of char formation upon combustion. For this reason, the use of lignin as a bio-based FR and antioxidative additive for synthetic polymers has been investigated since the early 2000s. In addition to its char formation, its antioxidative activity can be attributed to its high reactivity towards free radicals. Song and coworkers [61] described the preparation of a chemically-modified lignin derivative. Commercially-available lignin (isolated from wheat straw) was grafted with nitrogen and phosphorus-containing compounds to improve the intumescent properties. This was achieved by first treating lignin with formaldehyde in the presence of NaOH (aq). This hydroxymethylated lignin 30 was then subjected to reaction with the 31 to form 32 (Figure 16). This material was characterized by IR spectroscopy and XPS.

Figure 16. Preparation of 32.

Before this material was used as an FR additive, its thermal stability was first tested. The TGA showed that unmodified lignin (control) started to decompose at ca. 250 °C. A maximum weight loss (Tmax) was exhibited at 405 °C and a char residue of ca. 41% was left behind at 600 °C. The TGA revealed 30 to have slightly higher stability with a char residue of ca. 51%. Modified lignin 32 exhibited an initial degradation (5% weight loss) at 240 °C, with a more complex degradation pathway, leaving behind almost 62% char residue. The ability of this material to act as an FR additive was then tested. Modified lignin 32 was incorporated into poly(propylene-co-ethylene) (PPE) samples using melt blending. The thermal stabilities of these composites were then analyzed using TGA. Virgin PPE (control) exhibited initial decomposition (Ti) and Tmax values of 324 and 418 °C and left no char residue. Incorporation of 20 and 30 wt % of 32 to PPE lead to increases in Ti (325 and 361 °C,

max respectively)Polymers 2019, 11 and, 224 T (479 and 483 °C, respectively). These higher values are related to the higher20 of 31 char formations. The sample containing 20 wt % 32 left behind ca. 12% char. This value was comparable to polymer blends with 30 wt % pure lignin, showing that the incorporation of P and N canthe thermalincrease stabilitythe thermal and stability char formation. and char Coneformation. calorimetry Cone calorimetry was used to was measure used to the measure FR properties the FR propertiesof these materials. of these materials. It was shown It was that shown virgin that PPE virgin burns PPE rapidly burnswith rapidly a time with to a ignitiontime to ignition (tign) of ( 49tign s,) 2 ofPHRR 49 s, ofPHRR 1350 kW/mof 1350 ,kW/m and THR2, and of THR 87.3 MJ.of 87.3 Incorporation MJ. Incorporation of 20 wt %of 3220 exhibitedwt % 32 exhibitedtign of 38 s,tign PHRR of 38 ofs, PHRR380 kW/m of 3802, andkW/m THR2, and of 74.2THR MJ of while74.2 MJ 30 while wt % PN-lignin30 wt % PN-lignin incorporation incorporation values were values 39, were 360, and 39, 360, 69.7. andBoth 69.7. cases Both revealed cases dramaticrevealed decreasesdramatic decrea in PHRRses (ca.in PHRR 70%decrease) (ca. 70% decrease) and THR. and THR. Hu et al. [62] [62] used a similar strategy for the preparation of synergistic FR systems based on chemically-modifiedchemically-modified lignin lignin PU PU foams foams and and ammonium ammonium polyphosphate polyphosphate 8. Wheat8. Wheat straw straw lignin lignin was first was convertedfirst converted to 30 to, which30, which was wasthen then treated treated with with phosphoryl phosphoryl chloride, chloride, followed followed by byethylene ethylene glycol glycol to formto form modified modified lignin lignin 33 33(Figure(Figure 17). 17 ).The The PU PU foams foams were were prepared prepared by by mixing mixing 3333 andand PFAPP (microencapsulated 8,, prepared prepared from reaction with phenol, formaldehyde and and 8) with methylene diphenyl diisocyante (MDI) and po polyollyol in various amounts. The The th thermalermal behavior of the resulting materials was then investigated.

Figure 17. Preparation of lignin 33.

The TGA showed that pure PU exhibited first decomposition at 250 ◦C followed by second stage at 460–670 ◦C with no char residue. Analysis suggests that incorporation of PFAPP and 33 improved the PU foam stability: initial decomposition was increased to 278, 285, and 289 ◦C for composites containing 10, 20, and 30% modified lignin, respectively. Furthermore, these composites exhibited much higher levels of char residue upon combustion (28.1, 29.2, and 42.7%, respectively). Flammability properties were tested by LOI and cone calorimetry. Incorporation of 10, 20, and 30% 33 increased the LOI values to 23, 23.5, and 24.5% from 20% for pure PU. The PHRR values were dramatically reduced from 401 KW/m2 (pure PU) to 229, 193, and 165 KW/m2 for these samples. This noted decrease was caused from the protective char formation that occurs in the presence of modified lignin and 8. The covalent modification of the PU foam is important since simply adding typical FR materials to PU foams can result in dramatic mechanical deterioration. This was further illustrated by Chen, Wan, and coworkers [63] in 2014 in which they described another method of PU foam preparation using a lignin-based phosphate–melamine compound (LPMC). This material was chosen because phosphate-melamines are traditional intumescent FR. Lignin (obtained from corn straw used in ethanol production) was first subjected to liquefaction with sodium lignosulfonate and H2SO4 in a mixed solvent system of glycerin and low-molecular weight (ca. 400 Da) polyethylene glycol (PEG) at 160 ◦C. AlCl4 was added, followed by polyphosphoric acid (PPA), to perform the esterification. This was followed by the addition of melamine. Upon isolation, PU foams of varying LPMC content (0, 5, 10, 15, 20%) were prepared by mixing LPMC with dibutyltin dilaurate (catalyst), water (blowing agent) and polyphenylmethane polyisocyanate (PMDI). The authors noted that the mechanical properties could be enhanced upon the inclusion of up to 15% LPMC. Both Young’s modulus and compression stress exhibited a two-fold increase (from 0.51 and 11.34 MPa to 1.46 and 31.56 Mpa, respectively). At higher levels, though, deterioration of mechanical properties was revealed. ◦ ◦ The TGA revealed pure PU foam to have a Tinitial value of 291 C and a Tmax of 358 C and left only ca. 14% char at 600 ◦C. Pure LPMC, on the other hand, displayed more complex degradation ◦ with two Tmax values (314 and 383 C, corresponding to the cleavage of the phosphate ester bond and charring, respectively) and left behind 53.3% char residue up to 800 ◦C. Incorporation of LPMC led to Polymers 2019, 11, 224 21 of 31

◦ a decrease in Tinitial with increasing content (from 287.6 to 282.0 C for 5 and 20% LPMC, respectively), first stage Tmax values similar to that of pure PU foam and second stage Tmax values of 406.1, 407.4, 408.2, and 408.5 ◦C (for 5, 10, 15, and 20% LPMC content). As expected, the char residue formation increased upon increasing LPMC content: 17.2, 22.2, 24.4, and 27.7 wt % for each sample. FR properties were measured as well. LOI values for pure PU and PU foams containing 5, 10, 15, and 20 wt % LPMC were 18.9, 23.4, 24.8, 26.7, 28.3%. Advantageously, samples containing 15 and 20 wt % LPMC reached V-1 rating without dripping in UL-94 tests. The use of covalently-modified lignin as an FR additive to other synthetic polymers has been reported as well. Recently, Bourbigot’s [64] laboratory showed that kraft lignin could easily be phosphorylated through reaction with phosphorus pentoxide (P2O5). The resulting material (P-Lig) was then used to prepare FR ABS composites. Thermogravimetric analysis (TGA) was utilized to study the thermal behavior of these materials. Under air it was revealed that pure ABS exhibited T5% and ◦ Tmax values of 325 and 405 C. Upon the inclusion of 30 wt % P-Lig, these values could be decreased to 306 and 355 ◦C. Char residue was shown to be relatively low: 1.6% for pure ABS and only a two-fold increase with 30 wt % P-Lig. This could be increased upon analysis under inert atmosphere (ca. 17.1%). Calorimetry was further employed to study the flammability of each sample. Pure ABS exhibited PHRR and THR values of 482 kW/m2 and 72 MJ/m2. Upon the incorporation of 30 wt % P-Lig these values were decreased to 202 kW/m2 and 58 MJ/m2, indicating higher flame retardance. Furthermore, evolution of potentially toxic gases could be dramatically reduced. For instance, inclusion of 30 wt % P-Lig resulted in a 47% decrease in CO2 generation, 30% decrease in CO generation, 47% decrease in NO formation, and 27% decrease in HCN evolution.

PolymersAs 2018 noted, 10, earlier,x FOR PEER Laoutid REVIEW and coworkers [38] showed that biodegradable PLA could be instilled21 of 30 with high FR properties through melt-blending with a combination of phytic acid and lignin. In 2016, the samesame group group illustrated illustrated the the use ofuse P andof P N-derivatized and N-derivatized lignin as lignin an FR foras an PLA FR [65 for]. The PLA modification [65]. The modificationof kraft and organosolvof kraft and lignin organoso waslv performed lignin was through performed esterification through esterification with phosphorus with phosphorus chloride to chlorideform intermediate to form intermediate34, followed by34 reaction, followed with by ammonium reaction with hydroxide ammonium to provide hydroxide modified to ligninprovide35 modified(Figure 18 lignin). 35 was 35 (Figure used to 18). prepare 35 was FR used PLA to composites prepare FR using PLA melt composites processing. using melt processing.

Figure 18. Preparation of 35.

The authors studied the thermal properties of the resulting composites using TGA. Kraft and ◦ organosolv lignin lignin samples samples decomposed decomposed to to leave leave 50 50 and and 48% 48% residue residue at at800 800 °C. C.The The incorporation incorporation of Pof and P and N groups N groups (35) ( 35increased) increased these these values values to 58 to and 58 60% and 60%for kraft for kraftand organosolv, and organosolv, respectively. respectively. This indicatesThis indicates that the that modified the modified materials materials could could potentia potentiallylly serve serve as FR as additives. FR additives. PLA PLA composites composites of 20 of 20 wt % lignin and 20 wt % 35 were then analyzed. Pristine PLA exhibited a 10% weight loss (T ) wt % lignin and 20 wt % 35 were then analyzed. Pristine PLA exhibited a 10% weight loss (T10%10%) at ◦ ◦ 330at 330 °C whileC while composties composties of ofkraft kraft and and organosolv organosolv lignins lignins were were 225 225 and and 275 275 °C,C, respectively. The inclusion of lignin promotes char formation in whichwhich PLA–ligninPLA–lignin composites left behindbehind ca.ca. 15% 15% ◦ residue at 600 °C.C. The PLA composites of chemically-treated lignin ( 35),), however, however, exhibited thermal properties similar to those of pure PLA, but promotes more char formation (ca. 13%) than PLA. Cone calorimetry waswas utilized utilized to studyto study the FRthe properties FR proper of theseties composites.of these composites. It was found It thatwas incorporation found that incorporationof un-modified of lignins un-modified reduced lignins the PHHR reduced values the of PLAPHHR by values 21 (kraft) of andPLA 33% by (organosolv),21 (kraft) and due 33% to (organosolv),protective char due formation. to protective Interestingly, char formation. in the Interestingly, case of modified in the lignins case of 35modified, PHHR lignins values 35 could, PHHR be valuesdramatically could reducedbe dramatically only in with reduced kraft lignin.only in For with both kraft kraft lignin. and organosolv For both kraft35 samples, and organosolv TTI could 35 be samples,increased TTI from could two-fold be increased (compared from to two-fold unmodified (com ligninpared composites) to unmodified to values lignin similar composites) to pristine to values PLA. similar to pristine PLA. This resistance to ignition is due to the presence of the ammonium groups. Furthermore, composites prepared from 35 samples reached V0 ratings in UL-94 burn tests. Other groups have used the excellent char-forming capability of lignin to prepare FR PLA as well. That same year, Cayla et al. [66] prepared an intumescent system from kraft lignin (LK) and ammonium polyphosphate 8. They utilized twin-screw extrusion to prepare FR PLA composites of varying LK and 8 content. Selected data for thermal behavior measured using TGA is summarized in Table 11. Samples containing 10 wt % 8 were impervious to ignition in UL-94 flammability tests and exhibited no melt dripping. Samples composed of 5 and 10 wt % LK and 5 and 10 wt % 8 exhibited V0 rating.

Table 11. Thermal properties of PLA composites.

Sample Tonset (°C) Tmax (°C) Char Residue at 500 °C (%) PLA 330.7 370.5 1.1 LK 232.1 357.8 43.5 8 334.6 580.7 82.4 PLA-LK5%-8% 327.7 372.3 10.5 PLA-LK10%-8% 320.0 362.0 14.3

Wood–plastic composites (WPC) have found wide applicability in many industrial settings. Because of its cost and abundance, polypropylene (PP) is often utilized in this regard [67,68]. However, these composites are highly flammable. Recently, Song and Yu et al. [69] were able to fabricate WPC containing FR additives based off of covalently-modified lignin. To begin, the authors subjected purified alkali lignin to reaction with 7 in the presence of formaldehyde to provide 36 which was then subjected to reaction with diethylphosphite 37 and formaldehyde (Figure 19). This material was then metalated with copper(II)acetate (Cu(OAc)2) to form 38. The reason for the last step is that it has been shown that the presence of metal ions can have some synergistic effects on FR properties

Polymers 2019, 11, 224 22 of 31

This resistance to ignition is due to the presence of the ammonium groups. Furthermore, composites prepared from 35 samples reached V0 ratings in UL-94 burn tests. Other groups have used the excellent char-forming capability of lignin to prepare FR PLA as well. That same year, Cayla et al. [66] prepared an intumescent system from kraft lignin (LK) and ammonium polyphosphate 8. They utilized twin-screw extrusion to prepare FR PLA composites of varying LK and 8 content. Selected data for thermal behavior measured using TGA is summarized in Table 11. Samples containing 10 wt % 8 were impervious to ignition in UL-94 ﬂammability tests and exhibited no melt dripping. Samples composed of 5 and 10 wt % LK and 5 and 10 wt % 8 exhibited V0 rating.

Table 11. Thermal properties of PLA composites.

◦ ◦ ◦ Sample Tonset ( C) Tmax ( C) Char Residue at 500 C (%) PLA 330.7 370.5 1.1 LK 232.1 357.8 43.5 8 334.6 580.7 82.4 PLA-LK5%-8% 327.7 372.3 10.5 PLA-LK10%-8% 320.0 362.0 14.3

Wood–plastic composites (WPC) have found wide applicability in many industrial settings. Because of its cost and abundance, polypropylene (PP) is often utilized in this regard [67,68]. However, these composites are highly flammable. Recently, Song and Yu et al. [69] were able to fabricate WPC containing FR additives based off of covalently-modified lignin. To begin, the authors subjected purified alkali lignin to reaction with 7 in the presence of formaldehyde to provide 36 which was then subjected to reaction with diethylphosphite 37 and formaldehyde (Figure 19). This material was then metalated with copper(II)acetate (Cu(OAc)2) to form 38. The reason for the last step is that it Polymers 2018, 10, x FOR PEER REVIEW 22 of 30 has been shown that the presence of metal ions can have some synergistic effects on FR properties (i.e.,(i.e., enhancing enhancing char char formation). formation). Wood–plasticWood–plastic composites, of of 76.5 76.5 wt wt % % PP PP and and varying varying 38 38content,content, werewere prepared prepared via via melt melt compounding. compounding.

FigureFigure 19.19. SynthesisSynthesis of metalated lignin, lignin, 3838. .

TheThe flammability flammability properties properties of of each each compositecomposite were tested tested using using UL-94 UL-94 burn burn tests. tests. Wood–plastic Wood–plastic compositescomposites containing containing no no lignin lignin hadhad nono ratingrating while composites containing containing 15 15 wt wt %% pristine pristine lignin lignin werewere rated rated at at V-2. V-2. Wood–plastic Wood–plastic compositescomposites consisting of of 10 10 wt wt % % 3838 waswas rated rated atat V-1, V-1, indicating indicating betterbetter flame flame retardance. retardance. ConeCone calorimetrycalorimetry was empl employedoyed to to further further study study the the flame flame retardance retardance of of thesethese materials. materials. Wood–plastic Wood–plastic compositescomposites with no added lignin lignin showed showed PHRR PHRR and and THR THR values values of of 595595 kW/m kW/m2 2and and 93.993.9 MJ/ m m2.2 Inclusion. Inclusion of of15 15wt wt% lignin % lignin decreased decreased these these values values to 580 to kW/m 580 kW/m2 and 85.32 and 85.3MJ/m MJ/m2 while2 while using using 15 wt 15 % wt 38 %further38 further diminished diminished these values these to values 470 kW/m to 4702 and kW/m 70.22 MJ/mand 70.22. Further, MJ/m 2. Further,WPC consisting WPC consisting of either of lignin either or lignin 38 resulted or 38 resultedin decreased in decreased smoke production. smoke production. For example, For the example, TSR 2 2 the(total TSR smoke (total smokeproduction production rate) for rate) WPC for with WPC no withlignin no was lignin 12.5 was m /m 12.5, while m2/m the2, addition while the of addition 15 wt % of lignin or 38 decreased the TSR to 10.9 and 9.17 m2/m2, respectively. 15 wt % lignin or 38 decreased the TSR to 10.9 and 9.17 m2/m2, respectively. These same groups later used similar procedures to prepare FR composites based on These same groups later used similar procedures to prepare FR composites based on polybutylene polybutylene succinate (PBS) [70]. In this report, 36 (Figure 19) was metalated with zinc(II)acetate succinate (PBS) [70]. In this report, 36 (Figure 19) was metalated with zinc(II)acetate (Zn(OAc)2) to (Zn(OAc)2) to form PNZn-lig 39. Analysis by TGA in air revealed that 39 produced higher char residues (30 wt % at 800 °C) than did pristine lignin (25 wt %), indicating some enhancement of the FR properties. Pure PBS matrix exhibited Tinit and Tmax values of 354 and 410 °C. Incorporation of only 2.5 wt % lignin increased the Tinit (365 °C), but not the Tmax. The increase in Tinit is due to the early formation of protective char. Although the incorporation of 10 wt % 39 resulted in earlier Tinit values, it did leave the most char upon decomposition at 500 °C (10%). Similar to their findings with WPC, cone calorimetry revealed that the inclusion of modified lignin resulted in higher levels of flame retardance. For instance, PBS exhibited PHRR and THR values of 500 kW/m2 and 18.8 kJ/g. The incorporation of only 2.5 wt % lignin decreased these values to 416 kW/m2 and 21.6 kJ/g while 2.5 wt % 39 incorporation exhibited values of 447 kW/m2 and 14.4 kJ/g. These values could be further decreased upon increasing 39 content. For example, at 10 wt % 39, PHRR and THR were 244 kW/m2 and 6.1 kJ/g. All composites of lignin and modified lignin ceased droplet formation during combustion and resulted in higher levels of char formation (10 wt % 39 left ca. 55% residue). Additionally, these materials served to inhibit smoke production. The PBS matrix showed a TSR value of 274 m2/m2 while the composite consisting of 10 wt % 39 was 125 m2/m2. Because of its low cost and abundance in nature, the use of lignin as a carbon feedstock has been recognized. In fact, it can be used to prepare numerous small molecules. Vanillin 40 (the common component of vanilla) is one of the processes that has been commercially utilized and is presumably much cheaper and efficient than extraction from other sources [71]. Epoxy resins are among some of the most industrially-important polymeric materials. They are used in a wide variety of applications, but can be limited by their high levels of flammability. It is for this reason that the development of FR epoxy resins is crucial. In 2017, Ma, Zhu, and coworkers [72] reported on the preparation of FR epoxy resins based off of chemically-modified vanillin (produced from lignin). The synthesis started with the functionalization of 40 with 41 or 42 to form either 43 or 44. This was followed by the addition of diethylphosphite 45 in the presence of ZnCl to afford either 46 or 47 (Figure 20). These products were then subjected to reaction with epichlorohydrin 48 followed by base to provide epoxides 49 and 50. Subsequent curing with 46 (DDM) at 200 °C lead to vanillin-based epoxy resins.

Polymers 2019, 11, 224 23 of 31 form PNZn-lig 39. Analysis by TGA in air revealed that 39 produced higher char residues (30 wt % at 800 ◦C) than did pristine lignin (25 wt %), indicating some enhancement of the FR properties. Pure PBS ◦ matrix exhibited Tinit and Tmax values of 354 and 410 C. Incorporation of only 2.5 wt % lignin increased ◦ the Tinit (365 C), but not the Tmax. The increase in Tinit is due to the early formation of protective char. Although the incorporation of 10 wt % 39 resulted in earlier Tinit values, it did leave the most char upon decomposition at 500 ◦C (10%). Similar to their findings with WPC, cone calorimetry revealed that the inclusion of modified lignin resulted in higher levels of flame retardance. For instance, PBS exhibited PHRR and THR values of 500 kW/m2 and 18.8 kJ/g. The incorporation of only 2.5 wt % lignin decreased these values to 416 kW/m2 and 21.6 kJ/g while 2.5 wt % 39 incorporation exhibited values of 447 kW/m2 and 14.4 kJ/g. These values could be further decreased upon increasing 39 content. For example, at 10 wt % 39, PHRR and THR were 244 kW/m2 and 6.1 kJ/g. All composites of lignin and modified lignin ceased droplet formation during combustion and resulted in higher levels of char formation (10 wt % 39 left ca. 55% residue). Additionally, these materials served to inhibit smoke production. The PBS matrix showed a TSR value of 274 m2/m2 while the composite consisting of 10 wt % 39 was 125 m2/m2. Because of its low cost and abundance in nature, the use of lignin as a carbon feedstock has been recognized. In fact, it can be used to prepare numerous small molecules. Vanillin 40 (the common component of vanilla) is one of the processes that has been commercially utilized and is presumably much cheaper and efficient than extraction from other sources [71]. Epoxy resins are among some of the most industrially-important polymeric materials. They are used in a wide variety of applications, but can be limited by their high levels of flammability. It is for this reason that the development of FR epoxy resins is crucial. In 2017, Ma, Zhu, and coworkers [72] reported on the preparation of FR epoxy resins based off of chemically-modified vanillin (produced from lignin). The synthesis started with the functionalization of 40 with 41 or 42 to form either 43 or 44. This was followed by the addition of diethylphosphite 45 in the presence of ZnCl to afford either 46 or 47 (Figure 20). These products were then subjected to reaction with epichlorohydrin 48 followed by base to provide epoxides 49 and 50. ◦ SubsequentPolymers 2018, 10 curing, x FOR with PEER46 REVIEW(DDM) at 200 C lead to vanillin-based epoxy resins. 23 of 30

Figure 20. Preparation of 4646,, 47,, 49,, andand 5050..

Upon isolations,isolations, these resins werewere thenthen subjectedsubjected toto UL-94UL-94 burnburn tests.tests. The LOI valuesvalues resinresin preparedprepared fromfrom46 46and and49 49( 49(49––4646)) and and50 50––4646were were 31.4 31.4 and and 32.8%, 32.8%, respectively. respectively. These These were were both both rated rated at V0at V0 while while control control resin resin (DGEBA- (DGEBA-46) exhibited46) exhibited no rating no rating and hadand anhad LOI an value LOI value of 24.6%. of 24.6%. Further, Further,49–46 and49–4650 and–46 left50– behind46 left behind considerable considerable char residue char residue upon combustion upon combustion (35.8 and (35.8 41.2%). and Analysis41.2%). Analysis by TGA revealedby TGA thatrevealed both 49that–46 bothand 4950––4646 andexhibited 50–46 lower exhibitedTinit valueslower comparedTinit values to compared DGEBA-46 to, asDGEBA- is consistent46, as withis consistent P-containing with FR. P-cont Becauseaining of theFR. increased Because charof the formation, increased49 –46charand formation,50–46 exhibited 49–46 higherand 50T–max46 valuesexhibited than higher control Tmax resin. values As expected,than control control resin. DGEBA-DDM As expected, underwent control DGEBA-DDM complete decomposition underwent ◦ undercomplete air whiledecomposition50–46 left under almost air 60% while char 50 residue–46 left atalmost 700 C.60% char residue at 700 °C.

2.7. β-Cyclodextrin Cyclodextrins are macrocyclic structures comprised of multiple glucose units; wherein those constructed of 6, 7, or 8 (known as α-, β-, or γ-cyclodextrin) are the most common. These compounds are synthesized via enzymatic degradation of starch and their existence has been known for quite some time [73]. Because of its smooth decomposition and high char yield, β-cyclodextrin (β-CD, structure shown in Figure 21) has attracted much interest as an FR additive [74–76].

Figure 21. Molecular structure of β-cyclodextrin (β-CD).

Alongi and coworkers [77] described the use of β-CD as a dual action FR additive in which it serves as an IFR material as well as a nanofiller. The authors described the preparation of β-CD “nanosponges” (NS) through cross-linking β-CD units through carbonate linkages using simple mechanical grinding. These porous systems serve the purpose of trapping low molecular weight phosphorus-containing compounds (P), acting as a synergistic FR additive. The P species used were triphenyl phosphite 51, triethyl phosphate 52, ammonium polyphosphate 8, dibasic ammonium phosphate 53, and diethyl phosphoramidate 54 (Figure 22). These NS–P complexes were prepared

Polymers 2018, 10, x FOR PEER REVIEW 23 of 30

Figure 20. Preparation of 46, 47, 49, and 50.

Upon isolations, these resins were then subjected to UL-94 burn tests. The LOI values resin prepared from 46 and 49 (49–46) and 50–46 were 31.4 and 32.8%, respectively. These were both rated at V0 while control resin (DGEBA-46) exhibited no rating and had an LOI value of 24.6%. Further, 49–46 and 50–46 left behind considerable char residue upon combustion (35.8 and 41.2%). Analysis by TGA revealed that both 49–46 and 50–46 exhibited lower Tinit values compared to DGEBA-46, as is consistent with P-containing FR. Because of the increased char formation, 49–46 and 50–46 Polymersexhibited2019 higher, 11, 224 Tmax values than control resin. As expected, control DGEBA-DDM underwent24 of 31 complete decomposition under air while 50–46 left almost 60% char residue at 700 °C.

2.7. β-Cyclodextrin-Cyclodextrin Cyclodextrins are macrocyclic structures comprised of multiple glucose units; wherein those constructed of 6, 7, or 8 (known as α-,-, β--,, or γ-cyclodextrin)-cyclodextrin) are the most common. These compounds are synthesizedsynthesized viavia enzymatic enzymatic degradation degradation of of starch starch and and their their existence existence has beenhas been known known for quite for somequite timesome [ 73time]. Because [73]. Because of its smooth of its smooth decomposition decomposition and high and char high yield, charβ-cyclodextrin yield, β-cyclodextrin (β-CD, structure (β-CD, shownstructure in shown Figure 21in )Figure has attracted 21) has muchattracted interest much as interest an FR additive as an FR [ 74additive–76]. [74–76].

Figure 21. Molecular structure of β-cyclodextrin (β--CD).CD).

Alongi andand coworkerscoworkers [ 77[77]] described described the the use use of βof-CD β-CD as a as dual a dual action action FR additive FR additive in which in which it serves it asserves an IFR as materialan IFR asmaterial well as as a nanofiller.well as a Thenanofiller. authors The described authors the described preparation the of preparationβ-CD “nanosponges” of β-CD (NS)“nanosponges” through cross-linking (NS) throughβ-CD cross-linking units through β- carbonateCD units linkagesthrough usingcarbonate simple linkages mechanical using grinding. simple Thesemechanical porous grinding. systems These serve theporous purpose systems of trapping serve the low purpose molecular of trapping weight phosphorus-containing low molecular weight compoundsphosphorus-containing (P), acting ascompounds a synergistic (P), FR acting additive. as a synergistic The P species FR usedadditive. were The triphenyl P species phosphite used were51, Polymerstriethyltriphenyl 2018 phosphate phosphite, 10, x FOR52 PEER 51, ammonium, REVIEWtriethyl phosphate polyphosphate 52, ammonium8, dibasic ammoniumpolyphosphate phosphate 8, dibasic53, andammonium diethyl24 of 30 phosphoramidatephosphate 53, and54 diethyl(Figure phosphoramidate 22). These NS–P complexes 54 (Figure were 22). preparedThese NS–P through complexes further were grinding prepared with variousthrough Pfurther species. grinding The EVA with composites various P withspecies. these The materials EVA composites were then with prepared these materials via melt-blending. were then Theprepared authors via studied melt-blending. the flammability The authors properties studied of thethe resulting flammability composites properties using of standard the resulting UL-94 compositestests. Pristine using EVA standard burned completelyUL-94 tests. and Pristine rapidly EVA and burned was not completely classifiable. and Ratings rapidly of and V2 couldwas not be classifiable.achieved upon Ratings the incorporation of V2 could be of achieved 5, 10, and upon 15 wt the % NS-incorporation52 and 15 wtof %5, 10, NS and complexes 15 wt % of NS-8, 5352, and 1554 .wt The % behaviorNS complexes of these of materials8, 53, and was54. The further behavior probed of these using materials cone calorimetry. was further Pristine probed EVA using exhibited cone calorimetry.HRR value ofPristine 582 kW/m EVA 2exhibited. This value HRR could value be of decreased 582 kW/m upon2. This the value addition could of be NS–P decreased complexes. upon theFor addition instance, of the NS–P HRR complexes. of 10 wt % NS-For 52instance,was 177 the kW/m HRR2 ofwhile 10 wt 15 % wt NS- %52 NS- was53 was177 kW/m 177 kW/m2 while2 and 15 wt54 was% NS- further53 was decreased 177 kW/m to2 151andkW/m 54 was2 .further The PHRRs decreased were to dramatically 151 kW/m2 affected. The PHRRs and could were bedramatically decreased affectedfrom 1509 and kW/m could2 (forbe decreased EVA) to 254 from kW/m 15092 kW/mfor 152 wt(for % EVA) NS-54 to. 254 kW/m2 for 15 wt % NS-54.

Figure 22. Molecular structures of P-species used.

Polymers 2019, 11, 224 25 of 31

A couple of years later, the same group utilized these NS materials as synergistic FR additives for PP, LLDPE, and PA6 [78]. The NS–phosphorus complexes were prepared using mechanical grinding of NS and either 52 or 8. The resulting complexes were incorporated into the polymer matrix through mixing with PP, LLDPE, or PA 6 at elevated temperatures. The TGA revealed the effect that the NS–P complexes played on the thermal stability of the materials. Pristine PP exhibited 10% weight loss ◦ ◦ temperature (Tonset10%) of 430 C. Incorporation of the NS (10 wt %) alone lowered this to 406 C. This effect is due to the hydroxyl groups catalyzing the decomposition of the polymer. Values of 424, 439 ◦C were obtained for PP composites of NS-52 (7 wt % NS, 3 wt % 52) and 8 (7.5 wt % NS, 7.5 wt % 8) complexes, respectively. A similar effect was observed for LLDPE. It was argued that, in general, the NS–P complexes offered enhanced stability of polyolefins through the dehydration of β-CD moiety through the generation of phosphoric acid. The generation of water favored char formation. PA6 ◦ ◦ exhibited a Tonset10% of 411 C. Composites with NS-52 (7 wt % NS, 3 wt % 52) showed values of 334 C and 368 ◦C for NS-52 complex of 10 wt % NS and 5 wt % 52. The stability of the PA 6 is different than the polyofefins due to decomposition of the polymer during phosphoric acid generation. Even though TGA hinted at enhanced stability for these materials, char residues at temperatures of up to 700 ◦C were relatively low; the highest value was 8.6 wt % for the PA 6 composite of NS-52 (10 wt % NS and 5 wt % 52). Cone calorimetry was again employed to explore the flame-retardant properties of these composites. Pristine PP exhibited PHRR and THR values of 1541 kW/m2 and 90 MJ/m2. The values of composites of 7 wt % NS and 3 wt % TEP were 1529 kW/m2 and 93 MJ/ m2. These values could be improved to 839 kW/m2 and 90 MJ/m2 upon the incorporation of 10 wt % NS and 5 wt % 52. Composites of LLDPE exhibit different properties. The incorporation of both NS and NS–P complexes, in general, increased PHRR and THR. Incorporation of 15 wt % NS-52 or 8 lowered these values, but produced substantial amounts of smoke. Unfortunately, incorporation of these composites did not result in any observable FR properties to PA6. Considering that metal oxides and salts influence polymer degradation and char formation [79], Zhu’s [80] laboratory explored their incorporation into β-CD for FR polymer additives. The studied the effect such complexes have on the thermal stability and FR properties of poly(vinyl alcohol) (PVA), an important industrial polymer. The synthesis of the metalated CD (MC) complexes involved reaction of β-CD with maleic anhydride followed by formation of cross-linked metal complexes through the inclusion of hydroxide salts of Mg, Ca, and Ba to form MC–Mg, MC–Ca, and Mg–Ba, respectively. These synergists were incorporated into a PVA-8 mixture and samples were prepared using by solution casting. The authors first reported the UL-94 results for such systems. Ratings of V0 could be reached with 15 wt % FR and 0.1–0.5 wt % metal MC (Mg, Ca, or Ba) with 15 wt % 8 was suitable to obtain this rating. Analysis of the thermal stabilities by TGA revealed that incorporation of 8 and β-CD reduced the initial degradation temperatures (T5%) compared to PVA; this is due to the decomposition of 8. However, inclusion of metal salts increased this by approximately 16 ◦C; this is due to the stabilization of the metal ions. The composites containing metal ions also left behind higher levels of char residue. It was revealed that, although the initial degradation stage of these composites was lower than pristine PVA, the second and third stages were delayed. More recently, Vahabi, Shabanian, and coworkers [81] developed a new FR system based on β-CD, hydroxyapatite, and N-rich polymer, SABO 55, which is commercially available. The FR additive was prepared by first mixing β-CD, 55, and NaOH in DMF followed by addition of 56. This material (57) was subsequently reacted with hydroxyapatite (formed in situ from NH4PO6 and Ca(NO3)2 to form the FR additive 58 (Figure 23). Composites of 58-PLA (as well as 58-PLA-8) were prepared by melt mixing. Once prepared, thermal stabilities were studied using TGA. These analyses revealed that ◦ the incorporation of 58 slightly affected the T5% values, where pure PLA exhibited a value of 335 C while the PLA/58 was 334 ◦C. Inclusion of 8 further decreased this value to 323 ◦C. Whereas pure PLA decomposed to leave no char residue at 800 ◦C, pure 8 and 58 yielded 21 and 59%, respectively. PLA/58, PLA/8, and PLA/58/8 composites yielded values of 12, 15, and 11%, respectively. Although Polymers 2018, 10, x FOR PEER REVIEW 25 of 30

using by solution casting. The authors first reported the UL-94 results for such systems. Ratings of V0 could be reached with 15 wt % FR and 0.1–0.5 wt % metal MC (Mg, Ca, or Ba) with 15 wt % 8 was suitable to obtain this rating. Analysis of the thermal stabilities by TGA revealed that incorporation of 8 and β-CD reduced the initial degradation temperatures (T5%) compared to PVA; this is due to the decomposition of 8. However, inclusion of metal salts increased this by approximately 16 °C; this is due to the stabilization of the metal ions. The composites containing metal ions also left behind higher levels of char residue. It was revealed that, although the initial degradation stage of these composites was lower than pristine PVA, the second and third stages were delayed. More recently, Vahabi, Shabanian, and coworkers [81] developed a new FR system based on β- CD, hydroxyapatite, and N-rich polymer, SABO 55, which is commercially available. The FR additive was prepared by first mixing β-CD, 55, and NaOH in DMF followed by addition of 56. This material (57) was subsequently reacted with hydroxyapatite (formed in situ from NH4PO6 and Ca(NO3)2 to form the FR additive 58 (Figure 23). Composites of 58-PLA (as well as 58-PLA-8) were prepared by melt mixing. Once prepared, thermal stabilities were studied using TGA. These analyses revealed that the incorporation of 58 slightly affected the T5% values, where pure PLA exhibited a value of 335 Polymers°C while2019 ,the11, 224PLA/58 was 334 °C. Inclusion of 8 further decreased this value to 323 °C. Whereas pure26 of 31 PLA decomposed to leave no char residue at 800 °C, pure 8 and 58 yielded 21 and 59%, respectively. PLA/58, PLA/8, and PLA/58/8 composites yielded values of 12, 15, and 11%, respectively. Although 5858alone alone yielded yielded far far higher higher char char residue residue than than8 8,, thethe factfact that the the PLA/ PLA/5858 compositescomposites left left less less char char than than thethe PLA/ PLA/8 8composite composite is is not notdiscussed. discussed. In any case, this this data data is is indicative indicative of of possible possible enhancement enhancement of of FRFR properties. properties. To furtherTo further probe probe this, this, cone cone calorimetry calorimetry was employedwas employed to further to further study thestudy ﬁre the behavior, fire thebehavior, data is shown the data in is Table shown 12. in Table 12.

FigureFigure 23.23. Preparation of of 5858. .

Table 12. Thermal properties of PLA composites of 58 and 8. Table 12. Thermal properties of PLA composites of 58 and 8. 2 2 SampleSample PHRR PHRR (kW/m(kW/m2) )THR THR (MJ/m (MJ/m2) ) CharChar residue Residue (wt %) (wt %) PLAPLA 365365 89 89 0 0 PLA/58 365 73 5 PLA/58 365 73 5 PLA/8 302 65 15 PLA/8 302 65 15 PLA/58/8 185 45 40 PLA/58/8 185 45 40 As can be seen from the data, inclusion of 58 showed to have little effect on the fire behavior of PLA,As while can be 8 exhibited seen from modest the data, reduction inclusion in PHRR of 58 (fromshowed 365 to 301 have kW/m2). little effect A synergistic on the fire effect behavior was ofobserved, PLA, while however,8 exhibited with modest PLA composites reduction of in both PHRR 8 and (from 58, 365this tovalue 301 kW/m2).could be cut A synergisticalmost in half. effect wasFurther, observed, the THR however, values with were PLA also compositesreduced two-fo of bothld and8 andan increased58, this valuechar residue could be(40 cutwt almost%) was in half.observed. Further, the THR values were also reduced two-fold and an increased char residue (40 wt %) was observed.That same year, Fang and coworkers [82] described the covalent modification of β-CD with phenylThat samephosphonic year, Fang acid and dichloride coworkers 27 to [82 form] described new phospholidpidated the covalent modification β-CD 59 of (Figureβ-CD with 24). phenylThe phosphonicreasoning acidbehind dichloride this was27 theto formuse of new a phospholidpidatedP-containing β-CD wouldβ-CD 59favor(Figure direct 24 ).generation The reasoning of phosphoric acid upon combustion. Modified β-CD 59 was then used as an FR additive to PLA and as behind this was the use of a P-containing β-CD would favor direct generation of phosphoric acid a synergist in PLA-8 systems. Preparation of PLA composites were conducted through melt blending. upon combustion. Modified β-CD 59 was then used as an FR additive to PLA and as a synergist in The thermal stabilities of the samples were then analyzed using TGA. First, the authors compared PLA-8 systems. Preparation of PLA composites were conducted through melt blending. The thermal the thermal behaviors of 59 and β-CD. β-cyclodextrin alone exhibited T5% of 316 °C and left ca. 11 wt stabilities of the samples were then analyzed using TGA. First, the authors compared the thermal ◦ behaviors of 59 and β-CD. β-cyclodextrin alone exhibited T5% of 316 C and left ca. 11 wt % char ◦ ◦ residue at 700 C. Modified β-CD 59 exhibited a ca. 60 C decrease in T5% (due to decomposition of the P-moiety), but left behind app. 30 wt % char residue. This suggested that 59 may act as a competent FR ◦ additive. Composites of PLA and 59 were then studied. Pristine PLA exhibited a T5% value of 351 C. ◦ Incorporation of 30 wt % 59 lowered the T5% by 66 C. Additionally, this composite decomposed to leave higher char residue, albeit only 6.6 wt %. Synergistic effects of 59 with 8 were subsequently probed but first the authors found that composites of PLA with 8 (30 wt %) lowered the T5%% to 339 ◦C. Even though 8 promoted char formation, this char was unstable and further decomposed. The PLA composites with 30 wt % 8–59 (of varying ratios) led to increased char residue yields. Values close to 20 wt % could be achieved through the incorporation of 30 wt % 8–59 (2:1) and 30 wt % 8–59 (1:1). The authors deduced that this was from further esterification of hydroxyl groups of PCD with phosphoric acid generated from 8 decomposition. The fire behaviors of these systems were then studied. It was shown that the incorporation of 8 and 59 could have dramatic effects on LOI. These values could be increased from 19.7% (pure PLA) to 19.9% (PLA–59) and 33% (PLA–8), all the way up to 42.6% (PLA–8/59 (5:1). Furthermore, composites composed of 11:1, 5:1, 2:1, and 1:1 mixture of 8 and 59 were rated V0 in UL-94 tests. Polymers 2018, 10, x FOR PEER REVIEW 26 of 30

% char residue at 700 °C. Modified β-CD 59 exhibited a ca. 60 °C decrease in T5% (due to decomposition of the P-moiety), but left behind app. 30 wt % char residue. This suggested that 59 may act as a competent FR additive. Composites of PLA and 59 were then studied. Pristine PLA exhibited a T5% value of 351 °C. Incorporation of 30 wt % 59 lowered the T5% by 66 °C. Additionally, this composite decomposed to leave higher char residue, albeit only 6.6 wt %. Synergistic effects of 59 with 8 were subsequently probed but first the authors found that composites of PLA with 8 (30 wt %) lowered the T5%% to 339 °C. Even though 8 promoted char formation, this char was unstable and further decomposed. The PLA composites with 30 wt % 8–59 (of varying ratios) led to increased char residue yields. Values close to 20 wt % could be achieved through the incorporation of 30 wt % 8–59

Polymers(2:1) and2019 30, 11wt, 224% 8–59 (1:1). The authors deduced that this was from further esterification of hydroxyl27 of 31 groups of PCD with phosphoric acid generated from 8 decomposition.

Figure 24. Reported preparation of 5959.. 3. Conclusions The fire behaviors of these systems were then studied. It was shown that the incorporation of 8 and 59The could advent have and dramatic commercialization effects on LOI. of These synthetic values polymers could be has increased played from a tremendous 19.7% (pure role PLA) in advancingto 19.9% (PLA– our modern59) and lives. 33% The (PLA– explosive8), all growththe way of theup developmentto 42.6% (PLA– of new8/59 polymeric(5:1). Furthermore, materials, thatcomposites started composed in the twentieth of 11:1, century, 5:1, 2:1, showsand 1:1 nomixture signs of 8 slowing. and 59 were Continued rated V0 advances in UL-94 in tests. polymer science translate to advances in vital fields such as engineering and medicine, as well as luxuries like3. Conclusions cosmetics and apparel. However, there are downsides to further surrounding ourselves with theseThe synthetic advent materials, and commercialization one of which is of their synthetic relatively polymers high levels has played of flammability. a tremendous However, role in if theadvancing trend in our Figure modern1 is any lives. indication, The explosive discovering growth and of the inventing development ways toof preventnew polymeric this flammability materials, willthat alsostarted continue in the totwentieth grow. Furthermore, century, shows despite no signs repeated of slowing. environmental Continued blunders advances by thein polymer current administrationscience translate in to the advances United in States, vital fields the scientific such as engineering community and (as wellmedicine, as our as society well as in luxuries general) like is becomingcosmetics and more apparel. and more However, cognizant there of theare negativedownsides environmental to further surrounding effects of chemical ourselves research with these and production.synthetic materials, For this one reason, of which the implementationis their relatively of high the levels green of chemistry flammability. principles However, will continueif the trend to guidein Figure our 1 efforts is any in indication, developing discovering more sustainable and inventing FR. Paramount ways to to prevent this endeavor this flammability will be theincreased will also usecontinue of bio-based to grow. materials, Furthermore, which can despite be obtained repeated with lessenvironmental of a reliance onblunders finite petroleum by the reservescurrent thanadministration typical synthetic in the processes.United States, This rendersthe scientific their usecommunity inherently (as more well sustainable. as our society Selected in general) examples is ofbecoming bio-based more FR andand their more respective cognizant PHRR of the values negative discussed environmental in this review effects have of beenchemical shown research in Table and 13. Theproduction. examples For described this reason, in this the review implementation highlight many of the of thegreen positive chemistry discoveries principles that havewill continue been made to inguide this field.our efforts There isin stilldeveloping a long way more to go,sustainabl though.e One FR. detrimentParamount may to bethis incompatibility endeavor will between be the aincreased low-molecular use of bio-based weight, hydrophilic materials, which FR, and cana be hydrophobic obtained with polymer less of a that reliance could on potentially finite petroleum cause leachingreserves than of the typical FR. This synthetic problem processes. could be This solved rend throughers their covalent use inherently attachment more of sustainable. the FR in question, Selected butexamples could of also bio-based lead to changesFR and intheir the respective mechanical PH propertiesRR values of discussed the material. in this This review is obviously have been an issueshown that in Table will come 13. The up asexamples the field described progresses. in this Although review many highlight of the many bio-based of the FRpositive discussed discoveries can be obtainedthat havein been large made quantities, in this thisfield. is notThere always is still the a case.long way In fact, to thego, extractionthough. One of suchdetriment materials may can be solveincompatibility as a bottle between neck. More a low-molecular efficient, high throughputweight, hydrophilic technologies FR, and for obtaining a hydrophobic these materialspolymer willthat alsocould need potentially to be developed. cause leaching of the FR. This problem could be solved through covalent attachment of the FR in question, but could also lead to changes in the mechanical properties of the material. This Table 13. Selected examples of bio-based FR, polymeric material treated/prepared, and PHRR values. is obviously an issue that will come up as the field progresses. Although many of the bio-based FR discussedBio-Based can be obtained FR in Polymericlarge quantities, Material this is not PHRRalways the case. PHHR In fact, of Virginthe extraction Material of such materials cantannic solve acid as a bottle neck. epoxy More efficient, high335 kW/mthroughput2 technologies407 kW/m for obtaining2 these materials willphytic also acid need to be developed. EVA/chitosan 552 W/g 801 W/g isosorbide PLA 744 kW/m2 786 kW/m2 Tablediphenolic 13. Selected acid examples of bio-based PLA FR, polymeri388c material kW/m2 treated/prepared,418 and kW/m PHRR2 values. DNA EVA 963 kW/m2 1588 kW/m2 ligninBio-Based FR Polymeric PPE Material 405 PHRR kW/m 2 PHHR of Virgin1350 kW/m Material2 β-CD EVA 254 kW/m2 1509 kW/m2

Author Contributions: Writing-Original Draft Preparation, C.E.H.; Writing-Review & Editing, C.E.H. Funding: This research was funded by the National Science Foundation grant number 1744700 and the Sam Houston State University Department of Chemistry. Acknowledgments: C.E.H. would like to thank Diana Sepulveda (Hexion Corporation) for her insightful and invaluable conversations. Conﬂicts of Interest: The authors declare no conﬂict of interest. Polymers 2019, 11, 224 28 of 31

References

1. Hindersinn, R.R. Historical Aspects of Polymer Fire Retardance. In Fire and Polymers; Nelson, G.L., Ed.; ACS Symposium Series 425; American Chemical Society: Washington, DC, USA, 1990; pp. 87–96. ISBN 9780841212817. 2. Dickinson, E. The Complete Poems of Emily Dickinson; Johnson, T.H., Ed.; Back Bay Books, Little, Brown, and Co.: Columbus, GE, USA, 1997; p. 335. ISBN 0316184136. 3. Giraud, S.; Rault, F.; Cayla, A.; Salaun, F. History and Evolution of Fire Retardants for Textiles. In Proceedings of the COST MP1105 FLARETEX Workshop, Torino, Italy, 15–16 February 2016. 4. Muir, M.M.P. The Chemistry of Fire; Methuen: London, UK, 1893. 5. Oesper, R.E.M.M. Parrison Muir. J. Chem. Ed. 1945, 22, 458. [CrossRef] 6. U.S. Fire Problem. Available online: https://www.nfpa.org/News-and-Research/Data-research-and-tools/ US-Fire-Problem (accessed on 25 October 2018). 7. Kemmlein, S.; Herzke, D.; Law, R.J. Brominated Flame Retardants in the European Chemicals Policy of REACH-Regulation and Determination in Materials. J. Chromatogr. A 2009, 1216, 320–333. [CrossRef] [PubMed] 8. Horold, S. Phosphorus-based and Intumescent Flame Retardants. In Polymer Green Flame Retardants, 1st ed.; Papaspyrides, C.D., Kiliaris, P., Eds.; Elsevier: Amsterdam, The Netherlands, 2014; pp. 221–250. ISBN 0444538089. 9. Lu, S.-Y.; Hameron, I. Recent Developments in the Chemistry of Halogen-Free Flame Retardant Polymers. Prog. Polym. Sci. 2002, 27, 1661–1712. [CrossRef] 10. Kolb, V.M. Green Organic Chemistry and its Interdisciplinary Applications; CRC Press: Boca Raton, FL, USA, 2016. 11. Sonnier, R.; Taguet, A.; Ferry, L.; Lopez-Cuesta, J.-M. Towards Bio-based Flame Retardant Polymers; Springer: New York, NY, USA, 2018. 12. Xia, A.; Kiratitanavit, W.; Facendola, P.; Thota, S.; Yu, S.; Kumar, J.; Mosurkal, R.; Nagarajan, R. Fire Resistant Polyphenols Based on Chemical Modiﬁcation of Bio-Derived Tannic Acid. Polym. Degrad. Stab. 2018, 153, 227–243. [CrossRef] 13. Celzard, A.; Fierro, V.; Amaral-Labat, G.; Pizzi, A.; Torero, J. Flammability Assessment of Tannin-based Cellular Materials. Polym. Degrad. Stab. 2011, 96, 477–482. [CrossRef] 14. Grigsby, W.J.; Bridson, J.H.; Schrade, C. Modifuing Biodegradable Plastics with Additives Based on Condesnsed Tannin Esters. J. Appl. Polym. Sci. 2015, 132, 41626–41635. 15. Tributsch, H.; Fiechter, S. The Material Strategy of Fire-Resistant Tree Barks. In High Performance Structures and Materials IV, 1st ed.; de Wilde, W.P., Brebbia, C.A., Eds.; WIT Press: Southampton, UK, 2008; pp. 43–52. 16. Lang, X.; Shang, K.; Wang, Y.-Z.; Schiraldi, D.A. Low Flammability Foam-Like Materials Based on Epoxy, Tannic Acid, and Sodium Montmorillonite Clay. Green Mater. 2015, 3, 43–51. [CrossRef] 17. Moustafa, H.; Youssef, A.M.; Duquesne, S.; Darwish, N.A. Characterization of Bio-Filler Derived from Seashell Wastes and its Effect on the Mechanical, Thermal, and Flame Retardant Properties of ABS Composites. Polym. Compos. 2017, 38, 2788–2797. 18. Visakh, P.M. Advances in Flame Retardant of Different Types of Nanocomposites. In Flame Retardants: Polymer Blends, Composites and Nanocomposites, 1st ed.; Visakh, P.M., Arao, Y., Eds.; Springer: New York, NY, USA, 2015. 19. Wang, H.; Zhous, X.; Abro, M.; Gao, M.; Deng, M.; Qin, Z.; Sun, Y.; Yue, L.; Zhang, X. Mussel-Inspired General Interface Modiﬁcation Method and Its Application in Polymer Reinforcement and as a Flame-Retardant. ACS Omega 2018, 3, 4891–4898. [CrossRef] 20. Green, J. Mechanisms for Flame Retardancy and Smoke Suppression—A Review. J. Fire Sci. 1996, 14, 426–442. [CrossRef] 21. Xia, Z.; Singh, A.; Kiratitanavit, W.; Mosurkal, R.; Kumar, J.; Nagarajan, R. Unraveling the Mechanism of Thermal and Thermo-Oxidative Degradation of Tannic Acid. Thermochim. Acta 2015, 605, 77–85. [CrossRef] 22. Cho, J.H.; Vasagar, V.; Shanmuganathan, K.; Jones, A.R.; Nazarenko, S.; Ellison, C.J. Bioinspired Catecholic Flame Retardant Nanocoating for Flexible Polyurethane Foams. Chem. Mater. 2015, 27, 6784–6790. [CrossRef] 23. Salmeia, K.A.; Gaan, S.; Malucelli, G. Recent Advances for Flame Retardancy of Textiles Based on Phosphorus Chemistry. Polymers 2016, 8, 319. [CrossRef] Polymers 2019, 11, 224 29 of 31

24. Wang, B.; Shi, Y.; Hu, W.; Hong, N.; Zeng, W.; Ge, H.; Yu, X.; Song, L.; Hu, Y. Recent Advances for Microencapsulation of Flame Retardant. Polym. Degrad. Stab. 2015, 113, 96–109. [CrossRef] 25. Salmeia, K.A.; Gaan, S. An Overview of Some Recent Advances in DOPO-Derivatives: Chemistry and Flame Retardant Applications. Polym. Degrad. Stab. 2015, 113, 119–134. [CrossRef] 26. Chen, L.; Ruan, C.; Yang, R.; Wang, Y.-Z. Phosphorus-Containing Thermotropic Liquid Crystalline Polymers: A Class of Efficient Polymeric Flame Retardants. Polym. Chem. 2014, 5, 3737–3749. [CrossRef] 27. Lligadas, G.; Kozlowski, J.C.; Muzyczek, M.; Walentowska, J. Improving the Flame Retardancy of Plant Oil Based Polymers. In Polymer Green Flame Retardants, 1st ed.; Papaspyrides, C.D., Kiliaris, P., Eds.; Elsevier: Amsterdam, The Netherlands, 2014; pp. 221–250. ISBN 0444538089. 28. Marosi, G.; Szolnoki, B.; Bocz, K.; Toldy, A. Reactive and Additive Phosphous-Based Flame Retardants of Reduced Environmental Impact. In Polymer Green Flame Retardants, 1st ed.; Papaspyrides, C.D., Kiliaris, P., Eds.; Elsevier: Amsterdam, The Netherlands, 2014; pp. 221–250. ISBN 0444538089. 29. Duskova, D.; Marounek, M.; Brezina, P. Determination of Phytic Acid in Feeds and Faeces of Pigs and Poultry by Capillary Isotachophoresis. J. Sci. Food Agric. 2000, 81, 36–41. [CrossRef] 30. Zhang, T.; Yan, H.; Shen, L.; Fang, Z.; Zhang, X.; Wang, J.; Zhang, B. Chitosan/Phytic Acid Polyelectrolyte Complex: A Green and Renewable Intumescent Flame Retardant System for Ethylene-Vinyl Acetate Copolymer. Ind. Eng. Chem. Res. 2014, 53, 19199–19207. [CrossRef] 31. Zhang, T.; Yan, H.; Shen, L.; Fang, Z.; Zhang, X.; Wang, J.; Zhang, B. A Phosphorus-, Nitrogen- and Carbon-Containing Polyelectrolyte Complex: Preparation, Characterization and its Flame Retardant Performance on Polypropylene. RSC Adv. 2014, 4, 48285–48292. [CrossRef] 32. Horrocks, A.R. Developments in flame retardants for heat and fire resistant textiles—The role of char formation and intumescence. Polym. Degrad. Stab. 1996, 54, 143–154. [CrossRef] 33. Zheng, Z.; Liu, S.; Wang, B.; Yang, T.; Cui, X.; Wang, H. Preparation of a Novel Phosphorus- and Nitrogen-Containing Flame Retardant and its Synergistic Effect in the Intumescent Flame-Retarding Polypropylene System. Polym. Compos. 2015, 36, 1606–1619. [CrossRef] 34. Cheng, H.N.; Smith, P.B.; Gross, R.A. Green Polymer Chemistry: A Brief Review. In Green Polymer Chemistry: Biocatalysis and Materials II; Cheng, H.N., Gross, R.A., Smith, P.B., Eds.; ACS Symposium Series 1144; American Chemical Society: Washington, DC, USA, 2013; pp. 1–12. ISBN 9780841228955. 35. Haider, T.P.; Volker, C.; Kramm, J.; Landfester, K.; Wurm, F.R. Plastics of the Future? The Impact of Biodegradable Polymers on the Environment and on Society. Angew. Chem. Int. Ed. 2018, 57, 2–15. [CrossRef] [PubMed] 36. Tawiah, B.; Yu, B.; Fei, B. Advances in Flame Retardant Poly(Lactic Acid). Polymers 2018, 10, 876. [CrossRef] 37. Cheng, X.-W.; Guan, J.-P.; Tang, R.-C.; Liu, K.-Q. Phytic Acid as a Bio-Based Phosphorus Flame Retardant for Poly(Lactic Acid) Nonwoven Fabric. J. Clean. Prod. 2016, 124, 114–119. [CrossRef] 38. Costes, L.; Laoutid, F.; Brohez, S.; Delvosalle, C.; Dubois, P. Phytic Acid-Lignin Combination: A Simple and Efficient Route for Enhancing Thermal and Flame Retardant Properties of Polylactide. Eur. Polym. J. 2017, 94, 270–285. [CrossRef] 39. Xu, C.; Ferdosian, F. Conversion of Lignin into Bio-Based Chemicals and Materials; Springer: New York, NY, USA, 2017. 40. Nataraj, S.K.; Yang, K.S.; Aminabhavi, T.M. Polyacrylonitrile-Based Nanofibers-A State-of-the-Art Review. Prog. Polym. Sci. 2012, 37, 487–513. [CrossRef] 41. Ren, Y.; Zhang, Y.; Gu, Y.; Zeng, Q. Flame Retardant Polyacrylonitrile Fabrics Prepared by Organic-Inorganic Silica Coating via Sol-Gel Technique. Prog. Org. Coat. 2017, 112, 225–233. [CrossRef] 42. Richardson, J.J.; Bjornmalm, M.; Caruso, F. Technology-Driven Layer-by-Layer Assembly of Nanofilms. Science 2015, 348, 411–422. [CrossRef] 43. Jiang, Z.; Wang, C.; Fang, S.; Ji, P.; Wang, H.; Ji, C. Durable Flame-Retardant and Antidroplet Finishing of Polyester Fabrics with Flexible Polysiloxane and Phytic Acid Through Layer-by-Layer Assembly and Sol-Gel Process. J. App. Polym. Sci. 2018, 135, 46414–46424. [CrossRef] 44. Moulay, S. Chemical Modification of Poly(vinyl chloride)-Still on the Run. Prog. Polym. Sci. 2010, 35, 303–331. [CrossRef] 45. Hirschler, M.M. Poly(vinyl chloride) and its Fire Properties. Fire Mater. 2017, 41, 993–1006. [CrossRef] Polymers 2019, 11, 224 30 of 31

46. Cheng, L.; Wu, W.; Meng, W.; Xu, S.; Han, H.; Yu, Y.; Qu, H.; Xu, J. Application of Metallic Phytates to Poly(vinyl chloride) as Efficient Biobased Phosphorus Flame Retardants. J. Appl. Polym. Sci. 2018, 135, 46601–46611. [CrossRef] 47. Nelson, A.M.; Long, T.E. A Perspective on Emerging Polymer Technologies for Bisphenol—A Replacement. Polym. Int. 2012, 61, 1485–1491. [CrossRef] 48. Mauldin, T.C.; Zammarano, M.; Gilman, J.W.; Shields, J.R.; Boday, D.J. Synthesis and Characterization of Isosorbide-Based Polyphosphonates as Biobased Flame-Retardants. Polym. Chem. 2014, 5, 5139–5146. [CrossRef] 49. Daniel, Y.G.; Howell, B.A. Flame Retardant Properties of Isosorbide Bis-Phosphorus Esters. Polym. Degrad. Stab. 2017, 140, 25–31. [CrossRef] 50. Pileidis, F.D.; Titirici, M.-M. Levulinic Acid Biorefineries: New Challenges for Efficient Utilization of Biomass. ChemSusChem 2016, 9, 562–582. [CrossRef][PubMed] 51. Jing, J.; Zhang, Y.; Tang, X.; Fang, Z. Synthesis of a Highly Efficient Phosphorus-Containing Flame Retardant Utilizing Plant-Derived Diphenolic Acids and its Application in Polylactic acid. RSC Adv. 2016, 6, 49019–49027. [CrossRef] 52. Jing, J.; Zhang, Y.; Tang, X.; Li, X.; Peng, M.; Fang, Z. Combination of a Bio-Based Polyphosphonate and Modified Graphene Oxide toward Superior Flame Retardant Polylactic Acid. RSC Adv. 2018, 8, 4303–4313. [CrossRef] 53. Alongi, J.; Carletto, R.A.; Di Blasio, A.; Cuttica, F.; Carosio, F.; Bosco, F.; Malucelli, G. Intrinsic Intumescent-Like Flame Retardant Properties of DNA-Treated Cotton Fabrics. Carbohydr. Polym. 2013, 96, 296–304. [CrossRef] 54. Alongi, J.; Carletto, R.A.; Di Blasio, A.; Carosio, F.; Bosco, F.; Malucelli, G. DNA: A Novel, Green, Natural Flame Retardant and Suppresent for Cotton. J. Mater. Chem. A 2013, 1, 4779–4785. [CrossRef] 55. Alongi, J.; Carletto, R.A.; Di Blasio, A.; Carosio, F.; Bosco, F.; Malucelli, G. DNA: Flame Retardant Properites of Ethylene Vinyl Acetate Copolymers Melt-Compounded with Deoxyribonucleic Acid in the Presence of a-Cellulose or b-cyclodextrins. Curr. Org. Chem. 2014, 18, 1651–1660. [CrossRef] 56. Alongi, J.; Di Blasio, A.; Cuttica, F.; Carosio, F.; Malucelli, G. Bulk or Surface Treatments of Ethylene Vinyl Acetate Copolymers with DNA: Investigation on the Flame Retardant Properties. Eur. Polym. J. 2014, 51, 112–119. [CrossRef] 57. Alongi, J.; Cuttica, F.; Carosio, F. DNA Coatings from Byproducts: A Panacea for the Flame Retardancy of EVA, PP, ABS, PET, and PA6? ACS Sustain. Chem. Eng. 2016, 4, 3544–3551. [CrossRef] 58. Isarov, S.A.; Lee, P.W.; Towslee, J.H.; Hoffman, K.M.; Davis, R.D.; Maia, J.M.; Pokorski, J.K. DNA as a Flame Retardant Additive for Low-Density Polyethylene. Polymer 2016, 97, 504–514. [CrossRef] 59. Wang, Z.; Liu, Y.; Li, J. Regulating Effects of Nitrogenous Bases on the Char Structure and Flame Retardancy of Polypropylene/Intumescent Flame Retardant Composites. ACS Sustain. Chem. Eng. 2017, 5, 2375–2383. [CrossRef] 60. Xiao, B.; Sun, X.F.; Sun, R.C. The Chemical Modification of Lignins with Succinic Anhydride in Aqueous Systems. Polym. Degrad. Stab. 2001, 71, 223–231. [CrossRef] 61. Yu, Y.; Fu, S.; Song, P.; Luo, X.; Jin, Y.; Lu, F.; Wu, Q.; Ye, J. Functionalized Lignin by Grafting Phosphorus-Nitrogen Improves the Thermal Stability and Flame Retardancy of Polypropylene. Polym. Degrad. Stab. 2012, 97, 541–546. [CrossRef] 62. Xing, W.; Yuan, H.; Zhang, P.; Yang, H.; Song, L.; Hu, Y. Functionalized Lignin for Halogen-Free Flame Retardant Rigid Polyurethane Foam: Preparation, Thermal Stability, Fire Performance and Mechanical Properties. J. Polym. Res. 2013, 20, 234–246. [CrossRef] 63. Zhu, H.; Peng, Z.; Chen, Y.; Li, G.; Wang, L.; Tang, Y.; Pang, R.; Khan, Z.J.H.; Wan, P. Preparation and Characterization of Flame Retardant Polyurethane Foams Containing Phosphorus-Nitrogen-Functionalized Lignin. RSC Adv. 2014, 4, 55271–55279. [CrossRef] 64. Prieur, B.; Meub, M.; Witterman, M.; Klein, R.; Bellayer, S.; Fontaine, G.; Bourbigot, S. Phosphorylation of Lignin to Flame Retard Acrylonitrile Butadiene Styrene (ABS). Polym. Degrad. Stab. 2016, 127, 32–43. [CrossRef] 65. Costes, L.; Laoutid, F.; Aguedo, M.; Richel, A.; Brohez, S.; Delvosalle, C.; Dubois, P. Phosphorus and Nitrogen Derivatization as Efficient Route for Improvement of Lignin Flame Retardant Action in PLA. Eur. Polym. J. 2016, 84, 652–667. [CrossRef] Polymers 2019, 11, 224 31 of 31

66. Cayla, A.; Rault, F.; Giraud, S.; Salaun, F.; Fierro, V.; Celzard, A. PLA with Intumescent System Containing Lignin and Ammonium Polyphosphate for Flame Retardant Textile. Polymers 2016, 8, 331. [CrossRef] 67. Yang, T.H.; Yang, T.H.; Chao, W.C.; Leu, S.Y. Characterization of the Property Changes of Extruded Wood-Plastic Composites During Year Round Subtropical Weathering. Constr. Build. Mater. 2015, 88, 159–168. [CrossRef] 68. Najaﬁ, S.K. Use of Recycled Plastics in Wood Plastic Composites-A Review. Waste Manag. 2013, 33, 1898–1905. [CrossRef][PubMed] 69. Liu, L.; Qian, M.; Song, P.; Huang, G.; Yu, Y.; Fu, S. Fabrication of Green Lignin-Based Flame Retardants for Enhancing the Thermal and Fire Retardancy Properties of Polypropylene/Wood Composites. ACS Sustain. Chem. Eng. 2016, 4, 2422–2431. [CrossRef] 70. Liu, L.; Huang, G.; Song, P.; Yu, Y.; Fu, S. Converting Industrial Alkali Lignin to Biobased Functional Additives for Improving Fire Behavior and Smoke Suppression of Polybutylene Succinate. ACS Sustain. Chem. Eng. 2016, 4, 4732–4742. [CrossRef] 71. Fache, M.; Boutevin, B.; Caillol, S. Vanillin, A Key-Intermediate of Biobased Polymers. Eur. Polym. J. 2015, 68, 488–502. [CrossRef] 72. Wang, S.; Ma, S.; Xu, C.; Liu, Y.; Dai, J.; Wang, Z.; Liu, X.; Chen, J.; Shen, X.; Wei, J.; et al. Vanillin-Derived High-Performance Flame Retardant Epoxy Resins: Facile Synthesis and Properties. Macromolecules 2017, 50, 1892–1901. [CrossRef] 73. Wimmer, T. Cyclodextrins. Ullmann’s Encyl. Indust. Chem. 2012, 11, 23–31. 74. Marchal, A.; Delobel, R.; Le Bras, M.; Leroy, J.-M.; Price, D. Effect of Intumescence on Polymer Degradation. Polym. Degrad. Stab. 1994, 44, 263–272. [CrossRef] 75. Huang, L.; Gerber, M.; Lu, J.; Tonelli, A.E. Formation of a Flame Retardant-Cyclodextrin Inclusion Compound and its Application as a Flame Retardant for Poly(ethylene terephthalate). Polym. Degrad. Stab. 2001, 71, 279–284. [CrossRef] 76. Wang, H.; Li, B. Synergistic Effects of β-Cyclodextrin Containing Silicone Oligomer on Intumescent Flame Retardant Polypropylene System. Polym. Adv. Technol. 2010, 21, 691–697. [CrossRef] 77. Alongi, J.; Poskovic, M.; Frache, A.; Trotta, F. Novel Flame Retardants Containing Cyclodextrin Nanosponges and Phosphorus Compounds to Enhance EVA Combustion Properties. Polym. Degrad. Stab. 2010, 95, 2093–2100. [CrossRef] 78. Alongi, J.; Poskovic, M.; Visakh, P.M.; Frache, A.; Malucelli, G. Cyclodextrin Nanosponges as Novel, Green Flame Retardants for PP, LLDPE, and PA6. Carbohydr. Polym. 2012, 88, 1387–1394. [CrossRef] 79. Hassan, M.A.; Shehata, A.B. The Effect of Some Polymeric Metal Chelates on the Flammability Properties of Polypropylene. Polym. Degrad. Stab. 2004, 85, 733–740. [CrossRef] 80. Feng, J.; Zhang, X.; Ma, S.; Xiong, Z.; Zhang, C.; Jiang, Y.; Zhu, J. Synthesis of Metallic Cyclodextrins and Their Use as Synergists in Poly(Vinyl Alcohol)/Intumescent Flame Retardant System. Ind. Eng. Chem. Res. 2013, 52, 2784–2792. [CrossRef] 81. Vahabi, H.; Shabanian, M.; Aryanasab, F.; Laoutid, F.; Benali, S.; Reza Saeb, M.; Seidi, F.; Kandola, B.K. Three in One: β-Cyclodextrin, Nanohydroxyapatite, and a Nitrogen-Rich Polymer Integrated into a New Flame Retardant for Poly(Lactic Acid). Fire Mater. 2018, 42, 593–602. [CrossRef] 82. Zhang, Y.; Han, P.; Fang, Z. Synthesis of Phospholipidated β-Cyclodextrin and its Application for Flame-Retardant Poly(Lactic Acid) with Ammonium Polyphosphate. J. Appl. Polym. Sci. 2018, 135, 46054–46064. [CrossRef]

© 2019 by the author. 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/).