Recent Developments in Flame Retardant Paper -Polypropylene (PP) is widely used in the construction, automotive industry and packaging due to its excellent chemical resistance, easy processing and low cost


Contents lists available at



Polymer Degradation and Stability

journal homepage:


word image 2729

Polymer Degradation and Stability 178 (2020) 10920


Synergistic fire retardant effect between expandable graphite and ferrocene-based non-phosphorus polymer on polypropylene

word image 713

Zhou Cheng , Duijun Liao , Xiaoping Hu *, Wenxiong Li , Changqiong Xie , Haijun Zhang , Wenxue Yang

State Key Laboratory of Environment-Friendly Energy Materials, School of Materials Science and Engineering, Southwest University of Science and

Technology, Mianyang, 621010, China

a r t i c l e i n f o

a b s t r a c t

Article history:

Received 21 January 2020

Received in revised form

14 April 2020

Accepted 18 April 2020 Available online 26 April 2020



Expandable graphite

Ferrocene-based non-phosphorus polymer

Synergistic fire retardant effect

The key point to enhance the flame retardancy of expandable graphite (EG) in polymeric materials is to weaken the excessive intumescence of EG during combustion. Herein, a kind of ferrocene-based polymer (PDPFDE) is introduced into the flame retardant polypropylene (PP/EG) system. The fire retardant performance of PP composites was investigated by limiting oxygen index (LOI), vertical burning (UL-94) and cone calorimeter (CC) tests. With 20 wt% EG and 5 wt% PDPFDE added, the sample (PP3) passed the UL-94 V-0 rating, and its LOI value increased from 17% to 28.8% compared with neat PP (PP0). In contrast to PP0, CC data showed that the PP3 had a significantly depressed the peak heat release rate (pHRR) by 82.7%, and the total heat release (THR) by 41.0%. The thermal stability of PP composites and the synergistic effect between EG and PDPFDE were also studied by TGA and XRD. SEM and Raman analysis of the char residues of PP composites after UL-94 tests indicate that PP3 can produce dense char residue with a high degree of graphitization.

© 2020 Elsevier Ltd. All rights reserved.


Polypropylene (PP) is widely used in the construction, automotive industry and packaging due to its excellent chemical resistance, easy processing and low cost [1,2]. For example, PP is one of the matrices for preparing wood-plastic composites (WPC) used as building materials where needs safety requirements [3,4]. However, its intrinsic inflammability and serious melt dropping cause fire hazard and release toxic gases such as CO and flammable gases, which pollute the environment and harm to life [5,6]. In order to settle these severe issues, flame retardants, especially halogen-free flame retardants (HFFRs), were usually added owing to the urgent demand of present environmentally friendly development [7,8]. Among them, intumescent flame retardant (IFR) systems have a very wide application to PP for its halogen-free, low toxicity, and low smoke, etc [5]. However, almost all of the traditional IFR systems are phosphorus-containing flame retardant. But we know that traditional organophosphorus flame retardants (PFRs) come from non-renewable phosphate rock [9,10] and

* Corresponding author.

E-mail address: (X. Hu). 0141-3910/© 2020 Elsevier Ltd. All rights reserved.

contaminate the surrounding environment through abrasion, volatilization and leaching [11]. Moreover, 25 wt% loading of common IFR usually cannot achieve UL-94 V-0 rating for polypropylene owing to its low flame retardant efficiency [12,13].

Expandable graphite (EG) is one type of efficient intumescent flame retardant additive by acting as blowing agent in polymeric materials [14]. Scheme 1 shows the expansion process of EG when heated [15]. During combustion, EG can exfoliate or expand along the C-axis of the crystal structure by hundreds of times owing to the intercalation agent generate gas by blowing reaction, as it formed a driving force to thrust the layers along C-axis of graphite sheet [16,17]. Ultimately, a “worm-like” structure with a large number of ordered graphite sheets can be formed under a certain temperature [16], and the rapidly intumescence of EG when heated will induce the disorder expansion, producing excessive intumescent and polycellular char layer, so-called “popcorn effect” [18,19]. But, the loose char structure can not exert efficiently a physical protective barrier effect.

Nowdays, many researchers have focused on integrating phosphorus-containing flame retardants with EG, aiming to eliminate the “popcorn effect” of EG and enhance flame retardancy [16,20]. Qi et al. discovered that EG modified by 9,10-dihydro-9oxa-10-phosphaphenanthrene-10-oxide (DOPO) can impart PP

word image 2730

Scheme 1. The expansion process of EG under high temperature [15].

better flame retardancy than EG or EG/DOPO used alone [16]. Wang et al. observed that the edges of char residue of EG were sealed after the encapsulation by magnesium hydroxide (MH) nanosheets, resulting in the flame-retardant of PU-EG@ MH being significantly improved [21]. To date, few papers report the combination of nonphosphorus polymeric flame retardants and EG in improving the fire retardancy.

Ferrocene ((C5H5)2Fe), an aromatic transition metal “sandwich compound” containing iron and two cyclopentadienide ligands, is cheap, less toxic and heat stable [22]. As a well-known flame quencher, it reduced the flame speed and owned a good smoke suppression effect at the same time [23,24]. Furthermore, the total mass of smoke particles dropped for the enhanced oxidation of smoke catalyzed by iron decomposing from ferrocene [23]. The ferrocene-based polymer has good thermal stability and promotes the formation of denser residue char owing to the ferrocene unit [25]. Shahram Mehdipour-Ataei et al. prepared a series of ferrocene-based polymer with good thermal stability and high char yield endowing better flame retardancy [26].

Here, the molecular structure of PDPFDE (Scheme 2) is consist of ferrocene, benzene ring and ethylenediamine. Its self-highcharring performance imparts PDPFDE to improve the flame retardancy of EG in PP, which have the ability to settle the “popcorn effect” of EG. Ferrocene not only as a rigid unit for good thermal stability but also as a source of Fe which depressed the smoke. Benzene ring and cyclopentadienide ligands act as the main source of carbon that participates in the generation of char residue. The amine group in ethylenediamine provide the N element and contribute to the formation of residue char. In our previous study [13,27,28], we do a lot of research on ferrocene and used it to flame retardant PP and epoxy resins (EP). Liao et al. synthesis PDPFDE whose molecular structure contains ferrocene and benzene ring that possesses good thermal stability and char formation ability

word image 2731

Scheme 2. The structure of PDPFDE [27].

and the presence of nitrogen element as a gas source generally improved the flame retardancy of the composites [27]. Wen et al. revealed that poly 10-(2,5-dihydroxyphenyl) 9,10-dihydro-9-oxa10-phospha Phenan-threne-10-oxide-1,10-ferrocene dimethyl ester (PFDCHQ) that containing ferrocene can restrict the generation of smoke for EP and exhibit better flame retardancy [28]. Li et al. reported that PDPFDE combined with traditional IFR system to flame retardant PP, it is concluded that there is a synergistic effect between PDPFDE and IFR [13].

In this article, maleic anhydride grafted polypropylene (PP-gMAH, GB612) acts as a compatibilizer, PDPFDE is used as a synergistic agent accompanied by EG in proportion to prepare PP/EG/ PDPFDE composites. The flame retardant properties of composites investigated by LOI test, UL-94 test, CC test. The char residue of composites after UL-94 was characterized by SEM and Raman spectroscopy. The synergistic flame retardant mechanism between PDPFDE and EG was studied by TGA and XRD. The results showed nicely that an appropriate loading amount of PDPFDE plays a crucial role in improving the flame retardancy of PP/EG.



Polypropylene (PP, T30S) was obtained from PetroChina Company Limited, LanZhou Branch of Oil. Expendable graphite (EG, average particle size: 300 mm, expansion multiple: 350) was purchased from Nanjing Xianfeng Nano Materials Technology Co., Ltd., China. Maleic anhydride grafted polypropylene (PP-g-MAH, GB612) was provided by Shenzhen Huixin Plastic Chemical Co., Ltd., China. PDPFDE is synthesized by the method in our previous work in the laboratory [27].

Preparation of PP/EG/PDPFDE composites

Before preparing PP/EG/PDPFDE composites, the raw materials were dried in a vacuum oven at 60 C for 12 h, then proportionally add raw materials to the mixer (SU-70B, Changzhou Suyan Technology Co., Ltd.) at a temperature of 180 C with rotor speed of 36 rpm for 10 min. After completely cooling to room temperature, the mixed material was crushed into particles by powerful plastic crusher (PC-230, Shanghai Kende Machinery Co., Ltd.). Finally, different test samples were prepared by hot pressing at 180 C of 10 MPa for 5 min in a flat vulcanizing machine (XH-406, Dongguan City Xihua Testing Machines Co., Ltd., China), followed by cold pressing at 10 MPa for 5 min. The detailed formula of PP/EG/ PDPFDE composites is shown in Table 1.


Thelimitingoxygenindex(LOI)wastestedusingJF-3oxygenindex tester (Nanjing Jionglei Instrument Equipment Co., Ltd.) according to ASTM D2863-97. The samples dimension is100.0 6.5 3.0 mm3.

The vertical burning test (UL-94) was measured by M607 type horizontal vertical burning tester (Qingdao Shanfang Instrument Co., Ltd., China) based on the standard ASTM D3801 with a spline dimension of 130.0 13.0 3.2 mm3.

Cone calorimeter (CC) test was performed with standard FTT cone calorimeter (British FTTL company) according to the standard ISO 5660-1 with a sample size of 100.0 100.0 3.0 mm3. The sheet dimensions were exposed horizontally to an external heat flux of 35 kW/m2. All the samples were measured in triplicate and averaged.

The morphologies of the char residue of various composites after UL-94 test were examined by an Ultra 55 (German Carl

Table 1

LOI and UL-94 data of PP/EG/PDPFDE composites.

Samples PP (wt%) PP-g-MAH (wt%) EG (wt%) PDPFDE (wt%) LOI (%) UL-94


t1a þ t2b








17.0 ± 0.1









22.6 ± 0.3









25.7 ± 0.1









28.8 ± 0.1









28.0 ± 0.2









27.1 ± 0.1









25.6 ± 0.2




a The extinguish time after the first 10s ignition. b The extinguish time after the second 10s ignition. c It means the flame burns to the fixture. d It means no rating.

zeissNTS GmbH) scanning electron microscope (SEM) with an accelerating voltage of 15 kV.

Thermogravimetric analysis (TGA) and derivatives curves (DTG) were tested by STA 449C Jupiter Synchronous thermal analyzer (Germany NETZSCH Instrument Manufacturing Co., Ltd.) to research the thermal degradation properties of flame retardants and various PP/EG/PDPFDE composites. During testing, a high purity airstream was continuously transmitted to the furnace at a flow rate of 50 ml/min and the temperature heated from room temperature to 700 C with a heating rate of 10 C/min. In N2, the mass of PP0, PP1, PP2, PP3, PP4, PP5 and PP6 are 4.560 mg, 4.558 mg, 4.570 mg, 4.496 mg, 4.439 mg, 4.689 mg and 4.593 mg, respectively. In Air, the mass of EG, PDPFDE, EG/PDPFDE are 1.090 mg, 4.140 mg and 1.130 mg, respectively. Due to the expansion characteristics of EG, we must lower the usage of EG and EG/PDPFDE to prevent blocking the instrument. And all the samples are fine powder.

Raman spectroscopy tests were carried out using inVia (Renishaw) laser Raman spectrometer with a scan range of 1000e2000 cm1.

X-ray diffraction (XRD) analysis was carried out over a 2q range of 3e80 with a DMAX1400 X-ray diffractometer (Rigaku Co., Ltd., Japan), and equipped with Cu-Ka radiation (l ¼ 0.154 nm, 40 kV and 30 mA). The scan rate is 4 deg/min. Before the XRD test, EG, PDPFDE, EG/PDPFDE, PP1 and PP3 were heated in a furnace from room temperature to 700 C under a nitrogen atmosphere and then kept for 60 min.

Results and discussion

Limiting oxygen index and vertical burning tests

As we know, LOI and UL-94 tests are the basic and intuitive ways to observe the flame retardant properties of polymeric materials. Table 1 summarized the LOI values and UL-94 rating for neat PP (PP0) and its composites, the real-time digital photos of UL-94 tests for partial samples are shown in Fig. 1. According to the results, the LOI value of PP0 is only 17% and UL-94 test is no rating. Also, Fig. 1 shows that the flame spreads rapidly after the ignition for PP0, and accompanied by a severe melt dripping phenomenon.

When the addition amount of flame retardant is 25 wt%, the LOI values of all the PP composites increase to a certain extent. Adding 25 wt% EG alone, the LOI value of PP1 increases to 22.6%, but failing to pass the UL-94 test. As shown in Fig.1, since EG rapidly expanded into char layers, the barrier prevents heat transfer between the flame zone and the combustion matrix, thereby delaying the pyrolysis of the PP to a certain extent [29e32]. On the other hand, the intumescent chars formed with a serious “popcorn effect” phenomenon, which leads to the formation of loose chars that cannot

word image 2732

Fig. 1. The video screenshots of PP composites in the course of the UL-94 tests.

effectivity act as a physical barrier to heat and oxygen, thus negatively affects the fire retardancy of PP composites [20].

The effects of PDPFDE with different contents on the flame resistance of PP/EG/PDPFDE blends are studied while maintaining the total loading amount unchanged. When replacement of EG by 2.5 wt% PDPFDE, the LOI value of PP2 increases to 25.7%. Based on Fig. 1, the flame of PP2 is much smaller than PP1, probably due to the catalytic charring effect of PDPFDE [27], which helps to generate more char layer on the sample’s surface, then reducing the flame spread rate and weakening “popcorn effect” of EG during combustion. However, less PDPFDE is not enough for producing sufficient compact char layer, then falling through UL94 rating.

After uniting 20 wt% EG with 5 wt% PDPFDE, PP3 can pass UL-94 V-0 rating, and meanwhile, it obtains an LOI value of 28.8%. Further, Fig. 1 demonstrates clearly that the PP3 is self-extinguishing after ignition, indicating a dense and high-quality char layer produced rapidly on the surface of the sample, then acting as a good barrier to heat and oxygen [33]. It is worth mentioning that the “popcorn effect” of EG is eliminated. When the loading of PDPFDE increases to 7.5 wt%, PP4 can also pass UL-94 V-0 rating, the selfextinguishing time go up to 1.3 s. Whereas adding PDPFDE 10 wt % or 12.5 wt%, the UL 94 rating decreases to V-1 or lower, and the “popcorn effect” of EG reoccur. It should be noticed that the LOI values and UL-94 ratings of both PP5 and PP6 tend to decrease with the increasing loading ratios of PDPFDE. Thus, the more PDPFDE content cannot lead to a better fire retardancy, there is optimal addition ration between PDPFDE and EG. In other word, there is a distinct synergistic fire retarding effect between EG and PDPFDE. By incorporating excessive PDPFDE, its catalytic degradation effect will play a dominant role and go against the formation of compact char layer, resulting in a decrease in fire retardancy of PP composites


Thermal stability of PP/EG/PDPFDE composites

Fig. 2 and Table 2 illustrate the TGA and DTG curves of PP and its composites under N2 atmosphere. As shown in Fig. 2 and Table 2, the initial thermal degradation temperature (T5%) of PP0 is 409 C, and it undergoes one-step degradation during the whole process of degradation with almost no residue char above 490 C that indicates the complete decomposition and flammable characteristic. As for PP1 that added EG alone, the T5% and T10% is 23 C and 20 C higher than PP0, respectively, because of the good thermal stability of EG. As shown clearly in DTG, PP0 and PP1 undergo one-step degradation, but from PP2 to PP6 all samples undergo two-step degradation. And for PP1 to PP6, T5% and T10% show the composites almost decomposes in advance with the increasing loadings of PDPFDE, which the peaks around 270-285 C in DTG changed in the same way. These results illustrate the presence of PDPFDE catalyzes the degradation of PP composites. Moreover, the Tmax of them is similar but all higher than PP0, because not only EG but also PDPFDE possess good thermal stability [27].

The highest char yield of PP1 is caused by higher addition of EG that owns high char yield in N2 atmosphere. But the “popcorn effect” originated from excessive EG during combustion that can cause poor quality of the char residue of PP1 [21]. Nevertheless, from PP2 to PP6, the char yield increase first and then decrease. Among them, PP3 and PP4 have a higher char yield of 18.05%,

19.11%, respectively. The results may be attributed to the fact that PDPFDE helps to endow the corresponding composite materials good thermal stability and a high char residue, owing to its rigid structure of ferrocene and benzene ring [27]. However, there is an apparent synergistic effect between PDPFDE and EG, this phenomenon may be explained by two factors: (1) excessive PDPFDE will produce more Fe-based compound at high temperature, then promoting further decomposition of char residue; (2) excessive EG leads to “popcorn effect”; yet the decreasing usage of EG will weaken the barrier protection effect.

It is worth noting that the residual char yield of PP3 and PP4 is lower than that of PP1, the LOI and UL-94 rating of PP3 are significantly preferable. This result demonstrates that a dense and highquality residue char is generated for PP3 and PP4, as shown in the next SEM and Raman analysis, which contributes to good fire retardancy. The above experimental results mean that excellent fire performance is related to both char yield and its microstructure simultaneously [13]. Furthermore, the rate of Tmax of PP3 and PP4 decreased compared with PP1.

To further understand the synergistic effect between EG and PDPFDE, the experimental and calculated TGA curves of EG/PDPFDE according to their respective mass ratio in PP3 are comparatively analyzed. The calculated (cal.) TGA curves were summed up by the experimental (exp.) TGA curves of the mixture ingredients weighted by their contents using the following equation [19,34]:


McalðTÞ¼ X XiMiðTÞ (1)


where Xi is the content of substance i and Mi is the TGA curve of substance i.

Seen from Fig. S1 and Table S1, the EG with a high char yield at 700 C is in accord with the previous literature [21]. The experimental value of the residue at 700 C of EG/PDPFDE mixture in the air is 15% higher than the calculated one. And the T10%, T20% and T30% of EG/PDPFDE (exp.) are all significantly higher than those of EG/ PDPFDE (cal.). The experimental TGA curve is always higher than those of separate EG and PDPFDE after 259 C. Especially, T30% of EG/ PDPFDE (exp.) is 218 C higher than that of EG/PDPFDE (cal.). The results showed the synergistic charring effect does exist between EG and PDPFDE, which may be the main reason in improving the fire retardancy of PP composites.

Fire behaviour by cone calorimetry

Cone calorimetry (CC) has been widely used to detect the combustion performance of polymeric materials [35]. The detailed

word image 2733

Fig. 2. TGA (a) and DTG (b) curves of PP and its composites in N2 atmosphere.

Table 2

TGA data of PP and its composites in N2 atmosphere.


T5% (oC)a

T10% (oC)b

Tmax (oC)c

Rate of Tmax (%/oC)

Char at 700 C (%)











































a The temperature at 5% weight loss. b The temperature at 10% weight loss. c The temperature at maximum weight loss rate.

data obtained from CC test, such as time to ignition (TTI), peak heat release rate (pHRR), time to peak heat release (t-pHRR), total heat release (THR), average effective heat of combustion (AEHC), fire performance index (FPI), fire growth rate (FIGRA), and char, are summarized in Table 3. Comparing with the TTI of 54 s for PP0, the TTI of the PP composites all decrease to a certain extent. This result may come from the higher thermal conductivity of EG, which accelerates the heat to be transferred to the interior of PP composites [36,37]. What’s more, the PDPFDE will catalyze the thermal decomposition of the PP matrix at the same time.

Fig. 3(a) shows the HRR curves of the PP0, PP1, PP3 and PP4. The PP0 burns quickly after ignition and demonstrate a rapid and fierce heat release with one sharp pHRR value of 860.6 kW/m2. The pHRR values of PP1, PP3 and PP4 are reduced to 183.4 kW/m2 (78.7%), 148.6 kW/m2 (82.7%) and 155.8 kW/m2 (81.9%), respectively. To sum up, the combustion rate of PP1 was significantly reduced for the incorporation of EG. It’s worth noting that HRR of PP3 decreases continuously for the introduction of PDPFDE and EG at the same time. The reason is probably attributed to the physical barrier of EG and the synergistic effect between EG and PDPFDE. Compared to PP1, t-pHRR of PP3 and PP4 are 34s and 57s in advance, respectively. This is because that the PDPFDE catalyzes the early thermal degradation of PP matrix when ignition, which is beneficial to generate compact char layer, then hinders the further combustion and reaches the pHRR quickly [33,38e40]. Meanwhile, this illustrates the synergistic effect between PDPFDE and EG on flame retardant PP. As shown in Fig. 3(b), the PP0 presents a rapid and fierce heat release process in about 430s with burning exhaustion, and the total heat released by combustion is as high as 151.7 MJ/m2. By contrast, the THR of PP1 is only 94.9 MJ/m2 (37.4%), and the THR of PP3 and PP4 are further decreased to 89.5 MJ/m2 (41.0%) and 85.8 MJ/m2 (43.4%), respectively. As for PP1, the decreasing THR is probably due to the physical barrier effect of EG compared to PP0. But the “popcorn effect” and size effect of EG will severely weaken the fire retardancy of EG in PP [20], leading to no rating achieved in UL-94 test. For PP3 and PP4, the synergistic effect between PDPFDE and EG results in lower THR values and V-0 ratings in UL-94 tests. The char residue of PP0 and its composites are presented in Fig. 3(c). There is almost no residual char remaining for PP0 after burned, while the residual char yields of PP1, PP3 and

Table 3

Cone calorimetric data of PP0 and PP composites.

PP4 increase to 31.0%, 35.4% and 34.0%, respectively, which are all higher than the amount of flame retardant added (25 wt%). Moreover, the char residues of PP3 and PP4 are higher than those of PP1 with loading EG alone, indicating that EG and PDPFDE have synergistic effects in promoting char-formation. In summary, PDPFDE, act as a synergist replacing the partial EG by 5 wt% and 7.5 wt%, do enhance the fire retardancy of PP3 and PP4 in contrast to PP1.

AEHC is defined as the average heat released per unit mass of material being volatilized during combustion. The AEHC values of PP composites are found to be lower than that of PP0 (Table 3) due to fuel dilution by introducing flame retardant. FPI and FIGRA have been introduced to assess the fire hazards of the PP composite. FPI indicates the ratio of TTI to pHRR, implying the fire risk of a material, a high value of FPI indicates a high fire safety [41]. FIGRA is always equalled to value of pHRR/t-pHRR, and the lower the FIGRA value, the higher the fire safety [42]. As shown in Table 3, compared with PP1, the values of FPI of PP3 and PP4 increase and FIGRA decrease, signifying that the addition of PDPFDE can increase the fire safety factor of PP/EG system.

In order to explain the synergistic fire retardant effect of EG and PDPFDE in PP more directly, the flame inhibition effect, the barrier and protective effect, and the charring effect of PP composites are calculated in Table 4 based on the corresponding formula and data of CC tests [43]. On the whole, there is hardly any flame inhibition, the barrier and protective effect and the charring effect of PP1 systems account for a significant dominant role. While the incorporation of PDPFDE in PP3, both the charring effect and the barrier and protective effect are further enhanced comparing with PP1. As the addition of PDPFDE increases, the synergistic flame retardant effect in PP4 is slightly weakened but still better than that of PP1. This phenomenon clearly identifies that there is an optimal ratio (20 wt% EG and 5 wt% PDPFDE), which can endow PP matrix with the best fire retardancy. This result is in keeping with the results of LOI and UL-94 tests [13].

Usually, a lot of smoke and toxic gases originate from polymeric materials during combustion, which harms to human and environment [44]. The curves of smoke production rate (SPR) and CO production for pure PP and its composites are presented in Fig. 4. It is found that the addition of EG results in an obvious reduction in the peak smoke production rate (pSPR) and CO production with


TTI (s)

pHRR (kW/m2)

t-pHRR (s)

THR (MJ/m2)

AEHC (MJ/kg)

FPIa (m s2/KW)

FIGRAb (KW/m2s)

Char (wt%)


54 ± 1

860.6 ± 28.0

233 ± 2

151.7 ± 1.3

41.7 ± 0.2

0.063 ± 0.002

3.66 ± 0.03



43 ± 1

183.4 ± 5.8

232 ± 18

94.9 ± 2.1

41.3 ± 0.0

0.228 ± 0.007

0.89 ± 0.05

31.0 ± 1.2


37 ± 1

148.6 ± 7.1

198 ± 29

89.5 ± 2.8

39.6 ± 0.7

0.249 ± 0.011

0.68 ± 0.10

35.4 ± 0.8


41 ± 0

155.8 ± 3.2

175 ± 25

85.8 ± 0.5

39.7 ± 1.4

0.265 ± 0.007

0.77 ± 0.11

34.0 ± 0.5

a b FPI ¼ TTI/pHRR.


word image 2734

Fig. 3. (a) HRR, (b) THR and (c) Mass of PP and PP composites during cone calorimeter test.

Table 4

Quantitative assessment of the fire retardant effects for PP composites.


Flame inhibition effect (%)a

Charring effect (%)b

Barrier and protective effect (%)c














Flame inhibition effect ¼ 1 – AEHCFRPP/AEHCPP0.


Charring effect ¼ 1 – TMLFRPP/TMLPP0. c Barrier and protective effect ¼ 1 – (pHRRFRPP/pHRRPP0)/(THRFRPP/THRPP0); FRPP(flame retardant PP) means PP1, PP3, or PP4; TML means total mass loss.

word image 2735

Fig. 4. (a) SPR and (b) CO curves of PP and its composites during cone calorimeter tests.

respect to PP0. Especially, the corresponding pSPR value of PP3 is the lowest among all the samples. However, excessive loading of PDPFDE in PP4 will accelerate the thermal decomposition of PP molecules, then leading to an increased pSPR value.

According to our previous study and related literature, because 1-actylcyclopentadiene (C7H6O), Fe nanoparticles, benzene ring and cyclopentene (C5H5) are released upon the thermal decomposition of PDPFDE at high temperature [27,36], accordingly Fe nanoparticles can catalyze monomer/dimers of propylene, benzene ring, C5H5 and aromatic compounds (indene derivatives, naphthalene derivatives and biphenyl derivatives) to form carbon nanoparticles via rearrangement and aromatization [27,45]. Then it forms a graphite structure, producing more char residue and lowering the production of smoke [40]. Moreover, ferrocene is regarded as an effective smoke suppression agent. The above analysis can explain why the CO yield of PP3 and PP4 keep on decreasing slightly compared to PP1, demonstrating that PDPFDE can enhance the smoke suppression of EG.

Charring behaviour analysis

Morphology of chars

Fig. 5 displays the morphologies of char residues after UL-94 tests ofPPcomposites.ItisnoticeableinFig. 5(a,a1)thatthecharresidueof PP1 with EG alone exhibits the typical “worm-like” char with expanded graphite flakes [46]. In addition, the char layers of PP1 shows a loose, discontinuous and brittle with many holes on the surface, the reason may be that the poor interfacial compatibility between EG and PP matrix [18], likewise the “popcorn effect” of EG whenheated.Thistypeofresidualcharcannotfunctionasaneffective shield to heat and oxygen, which is the main reason why PP1 cannot pass the UL-94 test [47]. In Fig. 5(b, b1), compared with PP1, “popcorn effect” is restrained to some extent owing to the addition of 2.5 wt% EG. Fig. 5(c, c1) and (d, d1) show the SEM micrographs of residual chars for PP3 and PP4, respectively. Obviously, PP3 and PP4 produce continuous, dense and hard residue chars due to loading an appropriate amount of PDPFDE. A reasonable assumption is that 1-

word image 714

Fig. 5. Scanning electron micrographs (SEM) of residual char after UL-94 tests of PP composites: (a,a1) PP1, (b,b1) PP2, (c,c1) PP3, (d,d1) PP4, (e,e1) PP5, (f,f1) PP6.

actylcyclopentadiene (C7H6O), Fe nanoparticles, benzene ring and cyclopentene (C5H5) can be originated from PDPFDE during combustion. As the burning temperature goes up, Fe nanoparticles are easily oxidized to Fe3O4 [40]. On the other hand, monomer/dimers of propylene, benzene ring, C5H5, indene derivatives, naphthalene derivatives and biphenyl derivatives are catalyzed by Fe nanoparticle to form carbon nanoparticles [27]. Carbon nanoparticles would not only deposit on the surface of iron nanoparticles or EG flakes forming Fecored carbon nanotube (Fe-CNT) containing Fe3C phase, but also aromatized into graphite structure [38,48e50]. Finally, all formed substances above can produce a high-viscosity melt, which fills and connectsthegapsbetweenEGflakes[47].Thencombiningthebarrier effect of EG flakes, a compact and continuous char layer is formed, ultimately eliminating the “popcorn effect”. The detailed discussion will be illustrated in the XRD analysis and fire retardant mechanism sections.

For PP5, although the char layer is relatively compact, there are many traces of EG flakes on the surface of the char layer. Seen from Fig. 5(f, f1), the char layer of PP6 appears many EG flakes with big gaps. The reason is deduced that excessive PDPFDE plays a major role in catalytic degradation, which is harmful to the crosslinking of PP melt products, thus deteriorates the synergistic effect between EG and PDPFDE with a bad residue char. Thus, adding a proper amount of PDPFDE really weakens the “popcorn effect” of PP/EG systems effectively, consequently improving the fire retardancy of PP composites.

Raman analysis of chars

Raman spectroscopy is a powerful tool to characterize the degree of graphitization of residual char of PP composites [51]. The D band represents the vibrations of sp3 hybridized carbon atoms of disordered graphite, the G band reflects the vibrations of sp2 hybridized carbon atoms in a 2D hexagonal lattice [52]. Thus the degree of graphitization of char residue can be evaluated by the ratio of ID and IG, the smaller the ID/IG ratio means the greater the degree of graphitization with fewer defects. From Fig. 6, the ID/IG values from PP1 to PP6 in turn are 0.606, 1.269, 0.160, 0.527, 0.827 and 1.690, respectively. The lowest ID/IG value of PP3 means the corresponding residual char has the highest degree of graphitization, demonstrating that the appropriate amount of PDPFDE can improve the quality of residual char of PP composite. However, the higher ID/IG value of PP6 illustrates that more PDPFDE could deteriorate the char residue, which is consistent with the LOI and UL-94 tests data, further confirming that the optimum equilibrium ratio exists between EG and PDPFDE.

XRD analysis of chars

The XRD analysis of the EG, PDPFDE and EG/PDPFDE (the same mass ratio in PP3) mixtures at different temperatures is presented in Fig. 7(a) and Fig. 7(b). For EG, there almost no difference at room temperature and after heated, and the peaks at 26.26, 55.12 are the characteristics of (002) and (004) lattice plane of graphite (JCPDS: 99-0057) [53]. For PDPFDE, the XRD pattern is obviously changed after heated. The peaks at 2q¼ 30.18, 35.64 and 62.98 correspond to (220), (311) and (440) of Fe3O4 (JCPDS: 88-0315) [54]. The peaks at 42.89, 43.82, 44.70, 45.05 and 45.96 are the characteristic of (121), (210), (022), (103) and (211) reflections of Fe3C (JCPDS: 76-1877) [55]. Also, it contains the phase of Graphite at (002) and (004) reflection. The EG/PDPFDE has the same peaks as PDPFDE, but showing lower intensity. Besides, comparative XRD analysis of residue char of PP1 and PP3 after heating are depicted in Fig. 7(c). Both of them have the Graphite peak, but PP3 shows weak Fe3O4 and Fe3C phases emerging at the same locations as EG/ PDPFDE system. In summary, the residual chars of PDPFDE, EG/ PDPFDE and PP3 have the same phases of Graphite, Fe3O4 and Fe3C, and PP3 generates more compact residue char than PP1 as shown in SEM. These results show PDPFDE in PP3 plays a key role in improving the fire retardancy.

Fire retardant mechanism

Based on the previous analysis, the possible fire retardant mechanism of the PP/EG/PDPFDE system could be presented as Scheme 3. For PP1, EG swells to expanded graphite flakes and forms “worm-like” structure due to the “popcorn effect” of EG, producing loose char layer, which cannot effectively act as a physical protective barrier [18].

word image 2736

Fig. 6. Raman spectra of residue chars after UL-94 test for PP composites: (a) PP1, (b) PP2, (c) PP3, (d) PP4, (e) PP5, (f) PP6.

word image 2737

Fig. 7. XRD patterns of (a) EG and PDPFDE in room temperature, (b) EG, PDPFDE and EG/PDPFDE after 700 C heated in furnace under N2 atmosphere, (c) PP1 and PP3 after 700 C

heated in furnace under N2 atmosphere.

word image 2738

Scheme 3. Schematic illustrations of the fire retardant mechanism of PP/EG/PDPFDE systems.

For PP3, EG can also expand at a relatively low temperature, and the thermal decomposition of PDPFDE can produce 1actylcyclopentadiene, Fe nanoparticles, cyclopentene and benzene ring [27]. Fe nanoparticles is oxidized by the oxygen that come from the hydroxyl and carboxyl groups of EG and PDPFDE structure to form Fe3O4 [40]. Fe3O4 can act as a catalyst to accelerate the crosslinking of the polymeric matrix, facilitating the formation of high-quality residue char [56,57]. At the same time, Fe nanoparticles can catalyze monomer/dimers of propylene, benzene ring, C5H5, indene derivatives, naphthalene derivatives and biphenyl derivatives into carbon nanoparticles via rearrangement and aromatization based on the C-C coupling reaction, then generating graphite structure [48e50]. On the other hand, Fe-CNTs containing Fe3C formed at high temperatures can also promote the chemical crosslinking reaction of PP radicals [8,57]. These cross-linked structures produce high-viscosity melt during combustion [58,59], next, the melt products will fill into the gaps and connect the expanded graphite flakes [40], thus eliminating the “popcorn effect”. Finally, a compact and hard char layer with high graphitization structure is obtained. This protective char layer containing Fe-CNTs can significantly prevent the diffusion of oxygen into PP and the migration of volatile decomposed products into the flame zone. So PDPFDE not only has excellent self-catalytic charring capability [27], but also acts as the reactive charring adhesive of expanded graphite to suppress the further expansion of EG. In a word, an obvious synergistic fire retardant effect does exist between EG and PDPFDE in condensed phase to exert better fire retardancy for PP composites.


In summary, the synergistic fire retardant effect between EG and PDPFDE on PP was comprehensively studied. There is an equilibrium ratio between EG and PDPFDE, and the appropriate loading amount of PDPFDE play a key role in eliminating “popcorn effect” and improving the flame retardance of the PP/EG system. Excessive (PP6) or less (PP2) PDPFDE will deteriorate the synergy. The better fire retardancy and lower pSPR values of PP3 demonstrate that introducing PDPFDE can not only improve the flame retardancy but also suppress the smoke-production of PP/EG composite. And XRD analysis further confirms the synergistic flame retardant effect between PDPFDE and EG. What’s more, Fe3O4, Fe nanoparticle and Fe-CNT can promote the crosslinking reaction of PP radicals and the rearrangement and aromatization of pyrolysis production via C-C coupling reaction, consequently, a denser, continuous char residues with a high degree of graphitization can be obtained under the appropriate loading ratio of EG and PDPFDE in PP. This research will be helpful to provide a low-cost, environmentally-friendly and efficient fire retardant system for polypropylene.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

CRediT authorship contribution statement

Zhou Cheng: Investigation, Validation, Formal analysis, Writing – original draft. Duijun Liao: Investigation, Methodology. Xiaoping Hu: Methodology, Supervision, Writing – review & editing, Funding acquisition. Wenxiong Li: Conceptualization, Methodology. Changqiong Xie: Resources. Haijun Zhang: Software, Conceptualization. Wenxue Yang: Software, Conceptualization.


This work was financially supported by the National Natural Science Foundation of China (51973178; 51673160), The Project of State Key Laboratory of Environment-Friendly Energy Materials, Southwest University of Science and Technology (18FKSY0220), and Postgraduate Innovation Fund Project by Southwest University of Science and Technology (19ycx0027).

Appendix A. Supplementary data

Supplementary data to this article can be found online at


  1. S. Bourbigot, J. Sarazin, T. Bensabath, F. Samyn, M. Jimenez, Intumescent polypropylene: reaction to fire and mechanistic aspects, Fire Saf. J. 105 (2019) 261e269.
  2. P.J. Wang, X.P. Hu, D.J. Liao, Y. Wen, T.R. Hull, F. Miao, Q.T. Zhang, Dual fire retardant action: the combined gas and condensed phase effects of azomodified NiZnAl layered double hydroxide on intumescent polypropylene, Ind. Eng. Chem. Res. 56 (2017) 920e932.
  3. M. Nikolaeva, T. Ka€rki, Influence of fire retardants on the reaction-to-fire properties of coextruded wood-polypropylene composites, Fire Mater. 40 (2016) 535e543.
  4. J.Z. Liang, Tensile and flexural properties of polypropylene composites filled with highly effective flame retardant magnesium hydroxide, Polym. Test. 60 (2017) 110e116.
  5. P.K. Huang, F. Wu, Y.Y. Pang, M.H. Wu, X.Q. Lan, H.B. Luo, B. Shen, W.G. Zheng, Enhanced dispersion, flame retardancy and mechanical properties of polypropylene/intumescent flame retardant composites via supercritical CO2 foaming followed by defoaming, Compos. Sci. Technol. 171 (2019) 282e290.
  6. S. Xu, M. Zhang, S.Y. Li, H.Y. Zeng, J.Z. Du, C.R. Chen, Y. Pan, Surface modification of phosphorus-containing hydrotalcite using rare-earth coupling agent and its application in polypropylene, Powder Technol. 342 (2019) 555e561.
  7. Z.M. Zhu, W.H. Rao, A.H. Kang, W. Liao, Y.Z. Wang, Highly effective flame retarded polystyrene by synergistic effects between expandable graphite and aluminum hypophosphite, Polym. Degrad. Stabil. 154 (2018) 1e9.
  8. P.J. Wang, D.J. Liao, X.P. Hu, N. Pan, W.X. Li, D.Y. Wang, Y. Yao, Facile fabrication of biobased P-N-C-containing nano-layered hybrid: preparation, growth mechanism and its efficient fire retardancy in epoxy, Polym. Degrad. Stabil. 159 (2019) 153e162.
  9. Z.B. Shao, Z.C. Tang, X.Z. Lin, J. Jin, Z.Y. Li, C. Deng, Phosphorus/sulfur-containing aliphatic polyamide curing agent endowing epoxy resin with wellbalanced flame safety, transparency and refractive index, Mater. Des. 187 (2020) 108417.
  10. M.M. Velencoso, A. Battig, J.C. Markwart, B. Schartel, F.R. Wurm, Molecular firefighting-how modern phosphorus chemistry can help solve the challenge of flame retardancy, Angew. Chem., Int. Ed. Engl. 57 (2018) 10450e10467.
  11. G. Malarvannan, C. Belpaire, C. Geeraerts, I. Eulaers, H. Neels, A. Covaci, Organophosphorus flame retardants in the european eel in flanders, Belgium:

occurrence, fate and human health risk, Environ. Res. 140 (2015) 604e610.

  1. B. Yang, Y.H. Chen, M.D. Zhang, G.W. Yuan, Synergistic and compatibilizing effect of octavinyl polyhedral oligomeric silsesquioxane nanoparticles in polypropylene/intumescent flame retardant composite system, Composites Part A 123 (2019) 46e58.
  2. W.X. Li, D.J. Liao, X.P. Hu, Z. Cheng, C.Q. Xie, Synergistic improvement of fire retardancy and mechanical properties of ferrocene-based polymer in intumescent polypropylene composite, Polym. Adv. Technol. 30 (2019) 2402e2413.
  3. X.L. Chen, J.L. Zhuo, W.K. Song, C.M. Jiao, Y. Qian, S.X. Li, Flame retardant effects of organic inorganic hybrid intumescent flame retardant based on expandable graphite in silicone rubber composites, Polym. Adv. Technol. 25 (2014) 1530e1537.
  4. C.L. Dai, C.L. Gu, B.C. Liu, Y.H. Lyu, X. Yao, H.N. He, J.C. Fang, G. Zhao, Preparation of low-temperature expandable graphite as a novel steam plugging agent in heavy oil reservoirs, J. Mol. Liq. 293 (2019) 111535.
  5. F. Qi, M.Q. Tang, N. Wang, N. Liu, X.L. Chen, Z.B. Zhang, K. Zhang, X. Lu, Efficient organic-inorganic intumescent interfacial flame retardants to prepare flame retarded polypropylene with excellent performance, RSC Adv. 7 (2017) 31696e31706.
  6. D.K. Chattopadhyay, D.C. Webster, Thermal stability and flame retardancy of polyurethanes, Prog. Polym. Sci. 34 (2009) 1068e1133.
  7. W. Xi, L.J. Qian, Z.G. Huang, Y.F. Cao, L.J. Li, Continuous flame-retardant actions of two phosphate esters with expandable graphite in rigid polyurethane foams, Polym. Degrad. Stabil. 130 (2016) 97e102.
  8. Z.H. Zheng, Y. Liu, L. Zhang, H.Y. Wang, Synergistic effect of expandable graphite and intumescent flame retardants on the flame retardancy and thermal stability of polypropylene, J. Mater. Sci. 51 (2016) 5857e5871.
  9. Z.H. Zheng, Y.H. Liu, B.Y. Dai, C.Y. Meng, Z.X. Guo, Fabrication of cellulosebased halogen-free flame retardant and its synergistic effect with expandable graphite in polypropylene, Carbohydr. Polym. 213 (2019) 257e265.
  10. Y.T. Wang, F. Wang, Q.X. Dong, W.J. Yuan, P. Liu, Y.F. Ding, S.M. Zhang, M.S. Yang, G.Q. Zheng, Expandable graphite encapsulated by magnesium hydroxide nanosheets as an intumescent flame retardant for rigid polyurethane foams, J. Appl. Polym. Sci. 135 (2018) 46749.
  11. D.R. Van Staveren, N. Metzler-Nolte, Bioorganometallic chemistry of ferrocene, Chem. Rev. 104 (2004) 5931e5985.
  12. K. Tian, Z.S. Li, S. Staude, B. Li, Z.W. Sun, A. Lantz, M. Alden, B. Atakan, Influence of ferrocene addition to a laminar premixed propene flame: laser diagnostics, mass spectrometry and numerical simulations, Proc. Combust. Inst. 32 (2009) 445e452.
  13. C. Hu, W.Z. Li, Q.Z. Lin, X. Zheng, H.B. Pan, Q.F. Huang, Impact of ferrocene on the nanostructure and functional groups of soot in a propane/oxygen diffusion flame, RSC Adv. 7 (2017) 5427e5436.
  14. P.F. Chen, K.K. Yan, J.X. Leng, F. Zhang, L. Jiao, Synergistic smoke suppression effect of epoxy cross-linked structure and ferrocene on epoxy-based intumescent flame-retardant coating, Plast. Rubber. Compos. 47 (2018) 258e265.
  15. S. Mehdipour-Ataei, S. Babanzadeh, New types of heat-resistant, flameretardant ferrocene-based polyamides with improved solubility, React. Funct. Polym. 67 (2007) 883e892.
  16. D.J. Liao, Q.K. Xu, R.W. McCabe, H.V. Babu, X.P. Hu, N. Pan, D.Y. Wang, T.R. Hull, Ferrocene-based nonphosphorus copolymer: synthesis, high-charring mechanism, and its application in fire retardant epoxy resin, Ind. Eng. Chem. Res. 56 (2017) 12630e12643.
  17. Y. Wen, Z. Cheng, W.X. Li, Z. Li, D.J. Liao, X.P. Hu, N. Pan, D.Y. Wang, T.R. Hull, A novel oligomer containing DOPO and ferrocene groups: synthesis, characterization, and its application in fire retardant epoxy resin, Polym. Degrad. Stabil. 156 (2018) 111e124.
  18. X.L. Chen, H. Wu, Z. Luo, B. Yang, S.Y. Guo, J. Yu, Synergistic effects of expandable graphite with magnesium hydroxide on the flame retardancy and thermal properties of polypropylene, Polym. Eng. Sci. 47 (2007) 1756e1760.
  19. B. Dittrich, K.A. Wartig, D. Hofmann, R. Mülhaupt, B. Schartel, Flame retardancy through carbon nanomaterials: carbon black, multiwall nanotubes, expanded graphite, multi-layer graphene and graphene in polypropylene, Polym. Degrad. Stabil. 98 (2013) 1495e1505.
  20. B. Dittrich, K.A. Wartig, D. Hofmann, R. Mülhaupt, B. Schartel, The influence of layered, spherical, and tubular carbon nanomaterials’ concentration on the flame retardancy of polypropylene, Polym. Compos. 36 (2015) 1230e1241.
  21. D. Hofmann, K.A. Wartig, R. Thomann, B. Dittrich, B. Schartel, R. Mülhaupt, Functionalized graphene and carbon materials as additives for melt-extruded flame retardant polypropylene, Macromol. Mater. Eng. 298 (2013) 1322e1334.
  22. T. Tang, X.C. Chen, H. Chen, X.Y. Meng, Z.W. Jiang, W.G. Bi, Catalyzing carbonization of polypropylene itself by supported nickel catalyst during combustion of polypropylene/clay nanocomposite for improving fire retardancy, Chem. Mater. 17 (2005) 2799e2802.
  23. Y. Zhang, X.N. Li, Z.P. Fang, T.R. Hull, A. Kelarakis, A.A. Stec, Mechanism of enhancement of intumescent fire retardancy by metal acetates in polypropylene, Polym. Degrad. Stabil. 136 (2017) 139e145.
  24. B. Schartel, T.R. Hull, Development of fire-retarded materials-Interpretation of cone calorimeter data, Fire Mater. 31 (2007) 327e354.
  25. H.B. Yin, X.N. Gao, J. Ding, Z.G. Zhang, Experimental research on heat transfer mechanism of heat sink with composite phase change materials, Energy Convers. Manag. 49 (2008) 1740e1746.
  26. B. Dittrich, K.A. Wartig, D. Hofmann, R. Mülhaupt, B. Schartel, Carbon black, multiwall carbon nanotubes, expanded graphite and functionalized graphene flame retarded polypropylene nanocomposites, Polym. Adv. Technol. 24 (2013) 916e926.
  27. H.L. Zhu, Y.J. Bai, H.Z. Cui, L. Liu, Facile synthesis of carbon nanotubes via low temperature pyrolysis of ferrocene, J. Cryst. Growth 404 (2014) 44e47.
  28. G. Keru, P.G. Ndungu, V.O. Nyamori, Nitrogen-doped carbon nanotubes synthesised by pyrolysis of (4-{[(pyridine-4-yl)methylidene] amino} phenyl) ferrocene, J. Nanomater. 2013 (2013) 1e7.
  29. J.H. Zhang, Q.H. Kong, D.Y. Wang, Simultaneously improving the fire safety and mechanical properties of epoxy resin with Fe-CNTs via large-scale preparation, J. Mater. Chem. 6 (2018) 6376e6386.
  30. T. Jiang, X.H. Feng, Q.W. Wang, Z.F. Xiao, F.Q. Wang, Y.J. Xie, Fire performance of oak wood modified with N-methylol resin and methylolated guanylurea phosphate/boric acid-based fire retardant, Construct. Build. Mater. 72 (2014) 1e6.
  31. Z. Hu, L. Chen, G.P. Lin, Y. Luo, Y.Z. Wang, Flame retardation of glass-fibrereinforced polyamide 6 by a novel metal salt of alkylphosphinic acid, Polym. Degrad. Stabil. 96 (2011) 1538e1545.
  32. J.Y. Wang, L.J. Qian, Z.G. Huang, Y.Y. Fang, Y. Qiu, Synergistic flame-retardant behavior and mechanisms of aluminum poly-hexamethylenephosphinate and phosphaphenanthrene in epoxy resin, Polym. Degrad. Stabil. 130 (2016) 173e181.
  33. X. Dong, R.X. Qin, S.B. Nie, J.N. Yang, C. Zhang, W. Wu, Fire hazard suppression of intumescent flame retardant polypropylene based on a novel Ni-containing char-forming agent, Polym. Adv. Technol. 30 (2019) 563e572.
  34. Z. Li, S.I. Montero Lira, L. Zhang, D.F. Exposito, V.B. Heeralal, D.Y. Wang, Bioinspired engineering of boron nitride with iron-derived nanocatalyst toward enhanced fire retardancy of epoxy resin, Polym. Degrad. Stabil. 157 (2018) 119e130.
  35. N.V. Gama, R. Silva, F. Mohseni, A. Davarpanah, V.S. Amaral, A. Ferreira, A. Barros-Timmons, Enhancement of physical and reaction to fire properties of crude glycerol polyurethane foams filled with expanded graphite, Polym. Test. 69 (2018) 199e207.
  36. H.F. Zhu, Q.L. Zhu, J. Li, K. Tao, L.X. Xue, Q. Yan, Synergistic effect between expandable graphite and ammonium polyphosphate on flame retarded polylactide, Polym. Degrad. Stabil. 96 (2011) 183e189.
  37. T. Tang, X.C. Chen, X.Y. Meng, H. Chen, Y.P. Ding, Synthesis of multiwalled carbon nanotubes by catalytic combustion of polypropylene, Angew. Chem., Int. Ed. Engl. 44 (2005) 1517e1520.
  38. Z.W. Jiang, R.J. Song, W.G. Bi, J. Lu, T. Tang, Polypropylene as a carbon source for the synthesis of multi-walled carbon nanotubes via catalytic combustion, Carbon 45 (2007) 449e458.
  39. J. Gong, J. Liu, D. Wan, X.C. Chen, X. Wen, E. Mijowska, Z.W. Jiang, Y.H. Wang, T. Tang, Catalytic carbonization of polypropylene by the combined catalysis of activated carbon with Ni2O3 into carbon nanotubes and its mechanism, Appl. Catal. Gen. 449 (2012) 112e120.
  40. A.C. Ferrari, D.M. Basko, Raman spectroscopy as a versatile tool for studying the properties of graphene, Nat. Nanotechnol. 8 (2013) 235e246.
  41. F.S. Al-Hazmi, G.W. Beall, A.A. Al-Ghamdi, A. Alshahrie, F.S. Shokr, W.E. Mahmoud, Raman and ellipsometry spectroscopic analysis of graphene films grown directly on Si substrate via CVD technique for estimating the graphene atomic planes number, J. Mol. Struct. 1118 (2016) 275e278.
  42. X.Y. Pang, Y. Tian, X.Z. Shi, Synergism between hydrotalcite and silicatemodified expandable graphite on ethylene vinyl acetate copolymer combustion behavior, J. Appl. Polym. Sci. 134 (2017) 44634.
  43. X. Wang, F. Pan, Z. Xiang, Q.W. Zeng, K. Pei, R.C. Che, W. Lu, Magnetic vortex core-shell Fe3O4@C nanorings with enhanced microwave absorption performance, Carbon 157 (2020) 130e139.
  44. J.Q. Sheng, Y. Zhang, L. Liu, B. Quan, N. Zhang, G.B. Ji, Optimizing electromagnetic wave absorption performance: design from microscopic bamboo carbon nanotubes to macroscopic patterns, J. Alloys Compd. 809 (2019) 151866.
  45. K. Yao, J. Gong, N.N. Tian, Y.C. Lin, X. Wen, Z.W. Jiang, H. Na, T. Tang, Flammability properties and electromagnetic interference shielding of PVC/graphene composites containing Fe3O4 nanoparticles, RSC Adv. 5 (2015) 31910e31919.
  46. J. Gong, R. Niu, X. Wen, H.F. Yang, J. Liu, X.C. Chen, Z.Y. Sun, E. Mijowska, T. Tang, Synergistic effect of carbon fibers and carbon nanotubes on improving thermal stability and flame retardancy of polypropylene: a combination of a physical network and chemical crosslinking, RSC Adv. 5 (2015) 5484e5493.
  47. H.B. Zhao, B.W. Liu, X.L. Wang, L. Chen, X.L. Wang, Y.Z. Wang, A flameretardant-free and thermo-cross-linkable copolyester: flame-retardant and anti-dripping mode of action, Polymer 55 (2014) 2394e2403.
  48. Y.P. Ni, W.S. Wu, L. Chen, X. Zhao, Z.H. Qin, X.L. Wang, Y.Z. Wang, How hydrogen bond interactions affect the flame retardancy and anti-dripping performances of PET, Macromol. Mater. Eng. 305 (2019) 1900661.

word image 2739


Received: 11 April 2019

Revised: 30 May 2019

Accepted: 30 May 2019

DOI: 10.1002/pat.468


word image 2740

Synergistic improvement of fire retardancy and mechanical properties of ferrocene‐based polymer in intumescent polypropylene composite

Wen‐Xiong Li | Dui‐Jun Liao

State Key Laboratory of Environment‐Friendly Energy Materials, School of Materials Science and Engineering, Southwest University of Science and Technology, Mianyang, China


Xiao‐Ping Hu, State Key Laboratory for

Environment‐Friendly Energy Materials,

School of Materials Science and Engineering,

Southwest University of Science and

Technology, Mianyang 621010, China.


Funding information

Longshan Academic Talent Research Support

Plan of Southwest University of Science and Technology, Grant/Award Numbers:

18LZX440 and 17LZX404; National Natural

Science Foundation of China, Grant/Award

Numbers: 51673160 and 51373140;

Postgraduate Innovation Fund Project by

Southwest University of Science and Technology, Grant/Award Number:

19ycx0009; The Project of State Key

Laboratory of Environment‐Friendly Energy Materials, Southwest University of Science and Technology, Grant/Award Number:



The ferrocene‐based polymer (PDPFDE) accompanied with traditional intumescent flame retardant (IFR) system (ammonium polyphosphate (APP)/pentaerythritol (PER) = 3/1, mass ratio) has been used as additive flame retardant in polypropylene (PP), aiming to lower the total loading amount. The thermal stability and fire retardant properties were investigated by thermogravimetric analysis (TGA), limiting oxygen index (LOI), vertical combustion (UL‐94), and cone calorimetry (CONE). The fire retardant mechanism was studied by scanning electron microscopy (SEM), Fourier transform infrared spectroscopy (FTIR), and Raman spectroscopy. The results showed that the PP1 with 25 wt% IFR only passed the UL‐94 V‐1 rating, but the PP6 loaded by 0.5 wt% PDPFDE and 22.5 wt% IFR possessed an LOI value of 28.5% and passed the UL‐94 V‐0 rating; the peak heat release rate (pHRR) and total heat release (THR) are decreased by 63% and 43%, respectively, compared with pure PP. In addition, the char residue of PP6 manifested a very compact and smooth surface, indicating a more effective barrier layer. Meanwhile, it was interesting that the addition of PDPFDE evidently improved the impact strength and elongation at break of PP/IFR composites.


ferrocene‐based polymer, fire retardancy, mechanical properties, polypropylene, synergistic effect

inevitably deteriorate the mechanical properties of the flame retardant PP composites.15-18

2402 © 2019 John Wiley & Sons, Ltd. Polym Adv Technol. 2019;30:2402–2413.

Polypropylene (PP) is of numerous applicability in packaging, construction, electronics, automotive, and chemical industries due to its low density, easy processing, chemical resistance, high‐cost performance, and good mechanical properties.1-3 However, its high flammability limits its application in fire‐required fields, where most manufacturers are looking for halogen‐free alternatives to replace small molecule halogenated flame retardants, which released from their host polymer, and have been shown to be persistent, bioaccumulative, and toxic (PBT).4-8 In recent years, a mounting number of environment‐friendly intumescent flame retardants (IFRs) have been widely used in polyolefin.9-14 However, the low efficiency and poor compatibility of IFRs lead to the high‐loading amount, To figure out these shortcomings, Wang et al added adenine (A), guanine (G), cytosine (C), and uracil (U) as gas sources into IFR flame retardant PP system, respectively, in order to improve the flame retardant efficiency of IFR. PP composites containing 17 wt% IFR and 1 wt% C or U can pass the UL‐94 V‐0 rating and achieve a limiting oxygen index (LOI) value of 27.9%.16 Chen et al synthesized a charring agent called PETAT and used it to synergize with ammonium polyphosphate (APP) to flame retardant PP. When APP/PETAT = 2/1 (mass ratio) and the total loading is 25 wt%, the PP composites can pass the UL‐94 V‐0 rating, and the LOI value reaches 30.3%. Moreover, the tensile strength is not further damaged, and the bending strength is remarkably improved.17

Ferrocene (Fc) has been reported as an effective smoke suppressant and suggested as a potential fire retardant due to its iron content, cyclopentadienyl ring (Cp ring), and its well‐characterized air, moisture, and thermal stability.19-21 It may improve fire safety by flame quenching, smoke suppression, or enhancing char formation. A ferrocene‐based oligomer can also improve the flame retardancy of polymer because of its thermal stability, thermo‐oxidative stability, and ability to form carbonaceous protective layers.22 Furthermore, the Fc‐containing polymer possesses good bonding properties like inorganic salts and can have excellent mechanical properties with no need for the additional reinforcing agents.23 During the last two decades, Fc or ferrocene‐based polymers are widely used in flame retardant epoxy resins (EP), polystyrene (PS), and polyamides (PA).24-26 Chen et al reported that Fc could significantly promote the formation of denser residue char in the IFR EP system.24 Zhou et al found that Fc decorated on the graphene nanosheets could reduce the fire hazard of PS and improved the thermal stability of the polymer.25 Mehdipour‐Ataei et al revealed that PA containing Fc units exhibit better heat resistance and flame retardancy.26 However, Fc or ferrocene‐based polymers have rarely been reported for IFR PP.

In our previous study,27 it has been found that ferrocene‐based polymer (PDPFDE), as depicted in Scheme 1, is of high self‐charring ability and can catalyze the early thermal decomposition and crosslinking of EP to form char layer. The molecular structure of PDPFDE imparts its three main functions: (a) benzene ring and 1, 3‐cyclopentadiene can take effect as carbon source; (b) The Fen+, originated from Fc at high temperature, can take part in the molecular cross‐linking of IFR and polymer28; (c) The units of ethylenediamine, as a flexible chain segment contain nitrogen elements, can not only serve as a gas source but also improve the toughness of the polymeric composites. Thus, a potential synergistic effect may exist between PDPFDE and IFR, which can improve the flame retardant efficiency of the IFR flame retarding PP system, accompanied by enhancing its toughness.

In this paper, the main objective is to lower the overall loading of flame retardant in PP. The PDPFDE and IFR were mixed first in appropriate proportions, and then various PP/IFR/PDPFDE composites were obtained by melt blending. The effect of PDPFDE content on the thermal properties, fire retardancy, and mechanical properties of PP/IFR systems was in detail studied. Moreover, the synergistic flame retardant mechanism of PDPFDE in PP/IFR composites was also discussed comprehensively.


2.1 | Materials

PDPFDE was prepared in our own laboratory according to our reported paper,27 and its photograph and structural formula are shown in Scheme 1. PDPFDE is one kind of yellow powder and screened by 200 mesh before process. Ammonium polyphosphate (APP, analytical purity) was purchased from Tianjin Kemiou Chemical Reagent Co, Ltd (China). Pentaerythritol (PER, analytical purity) was purchased from Chengdu Kelong Chemical Reagent Factory (Sichuan, China). IFR represents the formula of APP/PER = 3/1 (mass ratio), and the above reagents were used without further purification. PP (T30S) was provided by PetroChina Lanzhou Petrochemical Company (Gansu, China).

2.2 | Preparation of flame retardant PP composites

First, all the raw materials were placed in a blast drying oven and dried at 60°C for 8 hours. Then, APP, PER, and PDPFDE were weighed and placed in a mortar according to the formulation shown inTable 1, used after grinding evenly. The PP and the uniformly ground flame retardant were mixed first and then placed in an internal mixer (SU‐70B, Changzhou Suyan Science and Technology Co, Ltd, China) with a rotor speed of 36 rpm for mixing for 15 minutes (The temperature of the three zones of the mixer is set at 180°C) to obtain a uniformly mixed PP/IFR/PDPFDE composite. After being completely cooled, it was broken into particles by a plastic pulverizer (PC‐230, Shanghai Kende Machinery Co, Ltd, China). Finally, it was hot pressed at 180°C and 10 MPa for 10 minutes, afterward cold pressed under the same pressure for 5 minutes by a flat vulcanizing machine (XH‐406, Dongguan City Xihua Testing Machines Co, Ltd, China) to obtain the standard test strips of different specifications in different molds.

2.3 | Characterization

LOI test: The test was carried out according to ASTM D2863‐97 using a JF‐3 oxygen index tester (Nanjing Jionglei Instrument Equipment Co, Ltd, China), and the sample size is 100.0 × 6.5 × 3.0 mm3.

UL‐94 vertical burning test: The test was performed by the M607 type horizontal vertical burning tester (Qingdao Shanfang Instrument

word image 2741 SCHEME 1 The photograph and molecular structure of PDPFDE [Colour figure can be viewed at]

TABLE 1 LOI and UL‐94 test results of PP and its composites


PP, wt%

IFR, wt%


LOI, %



t1/t2, s







17.0 ± 0.2








27.0 ± 0.1








27.9 ± 0.3








27.6 ± 0.2








24.5 ± 0.2








24.8 ± 0.1








28.5 ± 0.2








27.5 ± 0.1








24.6 ± 0.1








24.3 ± 0.2




Abbreviations: IFR, intumescent flame retardant; LOI, limiting oxygen index.

Co, Ltd, China) according to ASTM D3801, and the sample size is 130.0 × 13.0 × 3.2 mm3.

Thermogravimetric analysis (TGA) test: An STA 449C Jupiter type thermal analyzer (the German NETZSCH Instrument Manufacturing Co, Ltd) was employed. Test conditions: powder sample, sample dosage 3 to 5 mg, N2 atmosphere, a heating rate of 10°C/min from room temperature to 700°C.

Cone calorimetry (CONE) test: The tests were conducted according to ISO 5660‐1 using an FTT standard cone calorimeter (FTTL, UK) under an external heat flux of 35 kW/m2, the size of the sample is 100.0 × 100.0 × 3.0 mm3. All samples were tested three times.

Scanning electron microscopy (SEM): The morphologies of residual char after vertical burning test were observed using a TM‐1000 scanning electron microscope from Hitachi Limited, Japan. The PP composites were cryogenically fractured after being immersed in liquid nitrogen for 30 minutes, and then the microscopic morphology of the brittle fractured section was observed using a TM‐4000 scanning electron microscope from Hitachi Limited, Japan.

Fourier transform infrared (FTIR) spectroscopy: The residual char and KBr were mixed and compressed by FW‐4 tablet press and tested in the range of 4000 to 400 cm−1 by the Frontier type instrument of American PerkinElmer Instrument Company.

Raman spectroscopy: The residual char was tested using an in via type laser Raman spectrometer from Renishaw, UK. Scanning range: 500 to 2500 cm−1.

Mechanical properties of PP composites: ETM105D 100KN microcomputer controlled electronic universal testing machine (Changchun Intelligent Instrument Equipment Co, Ltd, China) was used for the tensile test according to GB/T1040.3‐2006, the tensile rate was 20 mm/min, and the samples were dumbbell type. The PIT501J LCD type plastic impact test machine (Shenzhen Wance Test Equipment Co, Ltd, China) was utilized for the impact test according to GB/T1043.1‐2008, the samples were unnotched strips, and the size was 80.0 × 10.0 × 4.0 mm3. All samples were tested three times.


3.1 | LOI and UL‐94 test

The LOI and UL‐94 tests were undertaken on PP and its composites to study the synergistic flame retardant effect between PDPFDE and IFR. The relevant data are shown in Table 1. The PP0 has an LOI value of only 17.0% and burns rapidly after ignition, together with severe melt dripping. The LOI value of PP1 containing 25 wt% IFR increases to 27.0%, and an UL‐94 V‐1 rating is achieved. When replacement of IFR by 0.5 and 1.0 wt% PDPFDE, respectively, the LOI values of PP2 and PP3 are improved slightly, and both of them can pass UL‐94 V‐0 rating. However, when the content of PDPFDE increases to 2.0 wt%, the LOI value of PP4 drops significantly to 24.5%, and melt dripping occurs during the combustion process, failing to pass the UL‐94 test. The above results show that PDPFDE and IFR exhibit obvious synergistic flame retardant effects in the case of low loadings of PDPFDE. But excessive PDPFDE will accelerate the catalytic degradation of the polymer matrix,29,30 which is not conducive to the formation of an effective protective char layer on the surface of the sample. Hence, the appropriate loading amount of PDPFDE is particularly crucial to the flame retardance of the PP/IFR system.

In order to further reduce the total loading of flame retardant, the formulations of PP6 and PP7 are designed. Both PP6 and PP7 can pass the UL‐94 V‐0 rating with no melt dropping, and the LOI value has an increase for PP6. Whereas neither PP8 nor PP9 can pass the UL‐94 test, and their LOI values decrease prominently. It is noticeable that the LOI value of PP6 reaches 28.5%, which is higher than that of PP2, indicating the optimal synergistic flame retardant effect between 0.5 wt% PDPFDE and 22.5 wt% IFR, then resulting in the lowest total loading of 23 wt% to reach the UL‐94 V‐0 rating.

Why do PP4 with 23 wt% IFR and 2 wt% PDPFDE fail the vertical burning test, but PP6 with 22.5 wt% IFR and 0.5 wt% PDPFDE can reach the UL‐94 V‐0 rating? It is attributed to excessive PDPFDE, which promotes accelerated decomposition and cross‐linking between PP matrix and IFR. Too fast and too much cross‐linking structure leads to a rapid increase in the viscosity of the composite melt, which is difficult to flow freely, forming a discontinuous char layer after combustion and allowing heat and mass transfer and inflammable volatiles to escape. Obviously, a char layer with a coarser surface and many holes and cracks cannot function as an effective barrier during


Figure S1 is a set of video screenshots of PP and its typical composites during the UL‐94 test. It is clear that PP0 burns rapidly after ignition and generates severe melt dripping. After adding 25 wt% IFR, the burning rate of PP1 is remarkably reduced, and a high‐quality char layer is quickly produced on the surface of the burned PP1, which plays a key role in hindering further combustion of PP matrix. However, PP5 containing 23 wt% IFR has a slow char formation rate during combustion, and it is arduous to form an effective protective char layer timely, consequently, give rise to no self‐extinguish after withdrawing the fire. After adding 0.5 and 1.0 wt% PDPFDE, respectively, both PP6 and PP7 carbonize and self‐extinguish within 0.32 second right after being deviated from fire, as a result, passing UL‐94 V‐0 rating. Because of the strong catalytic cross‐linking ability of PDPFDE,27 the synergistic effect of PDPFDE and IFR can quickly generate compact char layer during combustion for PP6 and PP7. It is extremely worth noting that no common intumescent char layer is observed, which may be for the reason that the catalytic cross‐linking function of PDPFDE causes the high‐viscosity melt, thus making the swollen behavior difficulty.

3.2 | Thermal stability of PP/IFR/PDPFDE composites

The TGA curves and derivative thermogravimetry (DTG) curves of PP and its composites under nitrogen atmosphere are depicted in Figure 1, and the corresponding data are listed in Table 2. It can be seen from Figure 1 that the PP0 has only one thermal degradation stage between 400 and 500°C. Its initial decomposition temperature (T5%) is 409°C, the maximum thermal weight loss rate temperature (Tmax) is 459°C, and the yield of char residue after 500°C is only

1.4%. After adding IFR, the T5% of PP1 and PP5 occurs earlier than TABLE 2 TGA and DTG data of PP and its composites in N2


T5%, °C

T10%, °C

Tmax, °C

Rate of



Char Yield, %


500°C 600°C


PP0 409 424 459 3.1 1.4 1.4 1.4

































Note. T5% is the temperature of 5% of the weight loss. T10% is the temperature of 10% of the weight loss. Tmax is the temperature corresponding to the maximum weight loss rate.

Abbreviation: TGA, thermogravimetric analysis.

word image 2742

FIGURE 1 A, Thermogravimetric analysis (TGA) and B, DTG curves of PP and its composites in N2 [Colour figure can be viewed at]

PP0, but the Tmax is improved. This is because the IFR decomposes at a lower temperature and forms an insulating char layer, delaying the further decomposition of the PP matrix. Compared with PP1 and PP5, the T5% of PP6 is advanced, and Tmax is almost unchanged, which can be attributed to the earlier decomposition and crosslinking of IFR to form char layer under the catalytic action of Fen+ produced by PDPFDE.28 It is worth noting that the T10% of PP6 and PP7 are higher than PP1 and PP5, especially the T10% of PP7 is slightly higher than PP0. This is because the Fen+ produced from PDPFDE28,33 can catalyze the earlier cross‐linking of the APP, PER, and PP matrix to form more effective protective char layer, thereby improving the thermal stability of the PP7 composite. The char residue yields of the PP composites containing IFR and/or PDPFDE are largely increased compared with PP0 (see Table 2). Except for PP7, the residual char contents of other composites change a little after 500°C. Although PP7 has the highest amount of residue char before 600°C in contrast to other PP composites, its final residual char at 700°C is only 7.4%. This result indicates that excessive PDPFDE will promote the further decomposition of residue char during the higher temperature stage. Although the residual char yield of PP6 and PP7 at 700°C is lower than that of PP5, the flame retardancy of PP6 and PP7 is significantly better than that of PP5. This can be explained by the denser char layer formed by PP6 and PP7. A detailed analysis of the microstructure of the char layer will be discussed later, indicating that good flame retardancy is not only related to the char yield but also depends mainly on the microstructure of the char layer.16

To further demonstrate the synergistic effect between IFR and PDPFDE, TGA was adopted to analyze the thermal stability of PDPFDE, IFR, and IFR/PDPFDE in a N2 atmosphere. As shown in Figure 2, the corresponding data are listed in Table 3. Among them, the experimental (Exp) TGA curve of IFR/PDPFDE was tested according to the mass ratio of IFR and PDPFDE of 45:1, and the calculated (Cal) TGA curve of IFR/PDPFDE was calculated according to the same mass ratio as the Exp TGA curves of PDPFDE and IFR. Although the weight percentage of PDPFDE in IFR/PDPFDE is very low, the Exp TGA curve of IFR/PDPFDE is significantly different from the Cal

TGA curve of IFR/PDPFDE. The T5%, T10%, and T15% of IFR/PDPFDE (Exp) are apparently lower than those of IFR/PDPFDE (Cal), respectively. Especially for T15%, the T15% of IFR/PDPFDE (Exp) is more than 20°C lower than that of IFR/PDPFDE (Cal), but the T30% of

IFR/PDPFDE (Exp) is obviously 14.8°C higher than that of IFR/PDPFDE (Cal). This means that PDPFDE does accelerate the decomposition of IFR at lower temperatures caused by Fen+ originated from the advanced decomposition of PDPFDE and then catalyzes the cross‐linking of IFR to form more char at higher temperatures. It should be noted that the char yield of IFR/PDPFDE (Exp) at 700°C is much lower than that of IFR/PDPFDE (Cal). This phenomenon indicates that PDPFDE can catalyze the further decomposition of the formed char layer above 500°C, which is consistent with the TGA results of PP6 and PP7.

3.3 | CONE test

Aside from LOI and UL‐94 tests, the fire behavior of PP and its composites was further evaluated by cone calorimeter, and its corresponding data are shown in Figure 3 and Table 4. PP0 burns quickly with a single peak, typical of intermediately thermally thin burning behavior; on the contrary, samples containing IFR or/and PDPFDE show the thermally thick burning behavior, typical of a char forming material.34 Table 4 shows that the time to ignition (TTI) of the PP5 is advanced

word image 2743

FIGURE 2 Thermogravimetric analysis (TGA) curves of PDPFDE, intumescent flame retardant (IFR), and IFR/PDPFDE in N2 [Colour figure can be viewed at]




T5%, °


T10%, °


T15%, °


T30%, °


Char Yield, %


500°C 600°C


PDPFDE 246.1 313.1 405.5 591.0 75.2 68.7 58.7



























Note. The mass ratio of IFR and PDPFDE in IFR/PDPFDE is 45:1. T15% is the temperature of 15% of the weight loss. T30% is the temperature of 30% of the weight loss.

Abbreviations: IFR, intumescent flame retardant; TGA, thermogravimetric analysis.

by 14 seconds after the loading of 23 wt% IFR, caused by the earlier decomposition of IFR, following produce the polyphosphoric acid (PPA) to catalyze the thermal decomposition of the PP matrix. After adding PDPFDE, the TTI values of PP6 and PP7 are postponed by 13 and 17 seconds, respectively, compared with PP5. The reason may be that PDPFDE decomposes to release Fen+ and also provides a carbon source.27 The Fen+ can catalyze the thermal cross‐linking reaction of APP, PER, and PP matrix at high temperature to form an effective protective char layer, greatly enhancing the thermal stability of PP/IFR system. This result is consistent with the conclusion of LOI and UL‐94 tests.

Based on Table 4, the peak heat release rate (pHRR) and total heat release (THR) values of PP0 reach 861 kW/m2 and 152 MJ/m2, respectively. On the other hand, the pHRR and THR values of PP5 decrease to 386 kW/m2 and 103 MJ/m2, respectively, reduced by 55% and 32% compared with PP0. As for PP6, its pHRR and THR values are 316 kW/m2 and 87 MJ/m2, which are 63% and 43% lower than those of PP0, respectively. Nevertheless, there is an increase in pHRR and THR values of PP7 in comparison with PP6, the reason may be that excessive PDPFDE in PP7 promotes the rapid crosslinking of IFR and PP matrix in early combustion, resulting in a high melt viscosity and difficult flow, and it is difficult to form an integrated char layer quickly, then more heat released by PP7. After 300 seconds, with the formation of a more complete carbonaceous layer in PP7, its HRR value gradually decreases, even a little lower than that of PP6. Moreover, it is clear to see that the HRR curve of PP7 is always higher than that of PP6 during the plateau. It shows that the introduction of PDPFDE does enhance the flame retardant efficiency of IFR in PP, and there is an optimal content of PDPFDE to keep a balance between catalyzing cross‐linking charring and decomposing, which further confirms the synergistic flame retardant effect of IFR and PDPFDE in PP matrix.

Surely, the above conclusions can also be inferred from the results of the average value of effective heat of combustion (av‐EHC) of PP

word image 2744

FIGURE 3 A, Heat rate release (HRR), B, total heat release (THR), C, mass loss, D, CO, and E, CO2 curves of PP and its composites during cone calorimeter tests [Colour figure can be viewed at]

TABLE 4 Main cone calorimetric data of PP and its composites


TTI, s

pHRR, kW/m2

t‐pHRR, s

THR, MJ/m2

Av‐EHC, MJ/kg

FPI, m2s/kW

FIGRA, kW/m2s

Char Residue, wt%


55 ± 1

861 ± 28

233 ± 2

152 ± 1

41.7 ± 0.2

0.063 ± 0.002

3.69 ± 0.12



41 ± 2

386 ± 12

338 ± 13

103 ± 3

36.6 ± 0.4

0.107 ± 0.002

1.14 ± 0.07

10.6 ± 0.4


54 ± 2

316 ± 7

294 ± 10

87 ± 1

35.6 ± 0.2

0.173 ± 0.004

1.07 ± 0.01

15.4 ± 0.3


58 ± 2

336 ± 9

288 ± 7

90 ± 3

36.5 ± 0.6

0.173 ± 0.005

1.17 ± 0.06

14.0 ± 0.3

Abbreviations: Av‐EHC, average value of effective heat of combustion; FIGRA, fire growth rate index; FPI, fire performance index; pHRR, peak heat release

rate; THR, total heat release; TTI, time to ignition.

and its composites (Table 4). The av‐EHC value of PP6 is 35.6 MJ/kg, which is 14.6% lower than that of PP0 (41.7 MJ/kg). Not only that, the av‐EHC values of PP5 and PP7 are significantly higher than that of PP6. It shows that PP6 with 22.5 wt% IFR and 0.5 wt% PDPFDE has more outstanding synergistic flame retardant effect. Appropriate amount of synergist can promote the formation of more stable and dense char residue in the intumescent flame retardant PP composite, thereby sequestering oxygen and heat in the condensed phase, resulting in incomplete combustion of some of the flammable volatiles to release less heat.

The mass loss rate (MLR) curves are shown in Figure 3c. Compared with PP5, PP6 and PP7 possess more residual char owing to the formation of compact char layer by the addition of PDPFDE. Meanwhile, more PDPFDE will reduce char residue of PP7 and lower its fire retardant properties.

The fire performance index (FPI) and the fire growth rate index (FIGRA) are two key parameters for evaluating the fire safety performance of materials. FPI = TTI/pHRR, FIGRA = pHRR/t‐pHRR, and t‐pHRR is the time to reach pHRR. In general, higher FPI values and lower FIGRA values indicate a high fire safety factor and enough time to escape in the fire scenario.1,35 As can be seen from Table 4, the FPI values of PP6 and PP7 are 62% higher than that of PP5. The FIGRA values of PP5, PP6, and PP7 are not much different, and the FIGRA value of PP6 is the lowest. The above results indicate that the appropriate amount of PDPFDE really improves the fire safety performance of PP/IFR composites.

In order to quantitatively evaluate the flame retardant effects of the PP composites, the flame inhibition effect, the charring effect, and the barrier and protective effect were calculated on the basis of the data of the CONE, and the calculation formula36 and results are

TABLE 5 Quantitative assessment of the flame retardant effects for PP composites


Flame Inhibition

Effect, %


Effect, %

Barrier and Protective

Effect, %













Note. Flame inhibition effect = 1 − av‐EHCn/av‐EHC0; charring effect = 1 − TMLn/TML0; barrier and protective effect = 1 − (pHRRn/ pHRR0)/(THRn/THR0); 0 means PP0; n means PP5, PP6, or PP7; and TML means total mass loss.

Abbreviation: Av‐EHC, average value of effective heat of combustion.

shown in Table 5. Compared with the three flame retardant effects of PP5, the flame retardant effects of PP6 are significantly enhanced. With the increase of PDPFDE loading, the flame retardant effects of PP7 are slightly weakened but still better than those of PP5. It confirms that 22.5 wt% IFR and 0.5 wt% PDPFDE can endow PP matrix with the best flame retardant performance, which is consistent with the test results of LOI and UL‐94.

The volatile products released during the combustion of the material are critical for the analysis of the flame retardant mechanism. Figure 3 illustrates the release rate curves of CO and CO2 during the CONE test of PP and its composites. The release rates of CO in PP composites are lower than that of PP. Compared with PP5, PP6 and PP7 release more CO and less CO2 after the introduction of PDPFDE. This is because PDPFDE can accelerate the formation of carbonization layer on the surface by catalyzing cross‐linking, similar to the analysis in LOI and UL‐94 tests. Along with the formation of the protective char layer, the O2 is hindered from the PP matrix and flammable volatiles, resulting in incomplete combustion of partially flammable volatiles to produce more CO, and conversely lessen the release of CO2.

3.4 | Structure of carbonization layer

Digital photographs of char residue of PP and its composites after CONE tests are represented in Figure S2. Apparently, there is almost no char residue remained for PP0 after combustion. Since a swollen char layer is obtained for PP5, the surface is cracked, which cannot play the role of a shield to heat and oxygen. Although there is no observable intumescent char residue obtained for both PP6 and PP7, the surface of char layer is unbroken, flat, and dense. This result is attributed to three reasons: (a) PDPFDE decomposes to produce Fen+at the early stage after ignition; (b) these Fen+ further catalyze the thermal decomposition of APP to produce PPA, then they are dehydrated from the PER and themselves to obtain more residual char; (c) Moreover, the produced Fen+ may participate in the crosslinking reaction of PPA and oxidized PP molecules in the form of bridged bond, leading to a higher melt viscosity during combustion.37 The lower fluidity of polymer melt is not beneficial to wrap the volatile products to form swelling carbonization layer, but just form compact char layer. It needs to be emphasized that the high melt viscosity might be responsible to inhibit melt dripping for PP6 and PP7 during UL‐94 testing.

The microstructure of the exterior surface of residual char after combustion is also observed by SEM, the results are displayed in Figure 4. It can be clearly seen that the char residue of PP5 is cracked and fragile. Such residual char cannot function to isolate oxygen and heat, this is the main reason why PP5 has a melt dripping phenomenon in the UL‐94 test and cannot pass the UL‐94 test. Comparatively, the denser, continuous, and compact char residue is obtained for PP6 after combustion, which can provide a powerful physical barrier effect to the external heat and oxygen entering the underlying PP matrix and the internal flammable volatiles, hence effectively slowing down further combustion of PP matrix. However, there are still some microvoids and cracks distributed on the surface of the residual char of PP7. This phenomenon is due to the excessive PDPFDE, which would promote overmuch cross‐linking and a higher melt viscosity, so that going against the formation of intact char residue.

word image 2745

FIGURE 4 Different magnifications of scanning electron microscopy (SEM) micrographs of the char residues after the UL‐94 tests for PP composites

Aiming to better understand the charring mechanism of IFR/PDPFDE system in PP, the char residues of various PP composites after the UL‐94 test were analyzed by FTIR; the results are shown in Figure 5. It can be found that the chemical structures of residual char of PP5, PP6, and PP7 are analogous, but individual content is quite different. The peaks at 3128, 3118, or 3120 cm−1 on each spectrum are the characteristic stretching vibrations of NH groups.38 The absorptions at 2951, 2919, 2869, 2839, 1457, and 1376 cm−1 are the characteristic peaks of saturated CH structure. The peak at 1631 cm−1 represents the characteristic stretching vibration of CC from polyaromatic structure.39 The peak at

word image 2746

FIGURE 5 Fourier transform infrared spectroscopy (FTIR) spectra of char residues after the UL‐94 tests for PP composites [Colour figure can be viewed at]

1400 cm−1 means the characteristic peak of PN structure.37 The peak at 1260 cm−1 is assigned to the characteristic peak of O groups,39 and 1167 and 492 cm−1 are the characteristic peaks of POC structure,40 and 998 cm−1 is the characteristic peak of POP structure.37 To sum up, all char residues of PP5, PP6, and PP7 contain polyaromatic structure and a cross‐linked structure such as O, POC, and POP; in particular, the unburned saturated hydrocarbon groups (CH3 and CH2) remain. It is highly interesting that the characteristic absorption peaks of saturated CH of PP6 are prominently stronger than that of PP5 and PP7, verifying that more unburned PP molecules are kept in the char residue. All the analysis identify that the PP6 with 0.5 wt% PDPFDE can achieve the best catalytic charring effect and form a dense and compact char layer then prevent the further combustion of the PP matrix.

Raman spectroscopy is further employed to investigate the charring mechanism of IFR/PDPFDE system in PP; the testing results are shown in Figure 6. The ID/IG values of PP5, PP6, and PP7 are 1.77, 1.42, and 1.65, respectively. Usually, the smaller the ID/IG value, the higher the degree of graphitization of the carbon material, indicates the better quality of the char layer.38,41 Raman spectroscopy results show that PP6 can form higher graphitizing char residue after combustion, indicating that the catalysis charring effect of PDPFDE can improve the quality of char residue in PP/IFR composites. The increasing ID/IG value of PP7 further identifies that there is an optimal loading content of PDPFDE in PP/IFR system, which controls the optimum equilibrium point between catalyzing cross‐linking charring and decomposing of PP/IFR system.

3.5 | Flame retardant mechanism

According to the above characterization results and analysis, the possible char formation mechanism of PP/IFR/PDPFDE in the combustion process is shown in Scheme 2. For PP5 (Scheme 2a), the char formation reaction during combustion can be divided into two parts: (a) First, the APP decomposes during the heating process to produce PPA and ammonia gas; second, PPA is dehydrated with PER and itself to obtain phosphoric ester structures; and third, the phosphoric ester is thermally pyrolyzed at a higher temperature and cross‐linked to form a char layer37,42; (b) The PP is dehydrogenated and oxidized to form COOH and COH groups at a higher temperature. Next, the COOH groups undergo a dehydroxylation radical reaction to obtain carbonyl structures; the COH groups are phosphorylated to form phosphoric ester structures; and due to the dehydrogenation of APP, PP is also dehydrogenated to obtain double bond structures.43 Ultimately, all of these structures are thermally pyrolyzed and/or cross‐linked to yield char residue.37,44

It is well known that the fire is caused by a spark; however, the fire spreads widely aroused by the mix of flammable gaseous chemicals and melts polymer.45 For PP6 and PP7 (Scheme 2b), the most critical distinction is that the Fen+ produced by PDPFDE can rapidly react with PPA to generate a Fe‐coordinated bond; meanwhile, Fen+ also reacts with the hydroxyl groups on the PP molecular chain to obtain a three‐dimensional network cross‐linked structure. Therefore, it is beneficial to improve the graphitization degree of residual char. It is worth noting that Fen+ acts as a bridge to connect the IFR and PP molecules; thus, more PP molecules participate in the cross‐linked structure, which leads to the increased thermal stability and viscosity of melt.28,37,44,46 The results are in accord with the analysis in LOI and UL‐94 tests. Besides, Fen+ can not only catalyze the dehydration between APP and PER to obtain a phosphoric ester37 but also promote the aromatization of unsaturated hydrocarbon chains obtained by dehydrogenation of PP molecular chains. With the transfer of heat, the PP molecular chain produces some unsaturated hydrocarbon chains due to the dehydrogenation ability of APP, then, due to the catalytic effect of Fen+, the unsaturated hydrocarbon chain undergoes

word image 2747

FIGURE 6 Raman spectra of char residues for PP composites after the UL‐94 tests [Colour figure can be viewed at] SCHEME 2 Possible char formation mechanisms for A, PP5 and B, PP6 and PP7 during combustion [Colour figure can be viewed at

word image]

word image 2749

FIGURE 7 Different magnifications of scanning electron microscopy (SEM) micrographs of the freeze‐fractured surface for PP composites

word image 2750

FIGURE 8 A, Impact strength and tensile strength and B, stress‐strain curves of PP and its composites [Colour figure can be viewed at]

dehydrogenation, cyclic dehydrogenation and cyclization, and further aromatization and isomerization into the char layer.47 In summary, the improved flame retardancy of PP/IFR/PDPFDE composites mainly relies on the reduction of combustible volatiles and the improvement of melt viscosity, which originated from the rapid formation of highquality char layers with higher graphitization degree.

3.6 | Mechanical properties of PP/IFR/PDPFDE composites

In addition to excellent flame retardant properties, mechanical properties are also a pivotal indicator for evaluating the practical performance of PP composites. Figure 8A shows the impact and tensile strength results of PP and its composites, and the relevant data are listed in Table S1. The impact strength of PP0 is 23.50 kJ/m2. When IFR is added to PP, the impact strength values of PP1 and PP5 are decreased by 26.1% and 22.0%, respectively. This result is ascribed to the high loading and poor compatibility of IFR in PP matrix,48,49 which leads to the formation of stress concentration inside the PP matrix and a reduction in impact strength. At the same time, when 0.5 or 1.0 wt% IFR in PP5 is replaced by PDPFDE, the impact strength figures of PP6 and PP7 are improved by 29.5% and 30.4% compared with PP5. The SEM images of the freezefractured surface of PP composites are shown in Figure 7. For PP5, the fractured surface is rough and accompanied by large voids; moreover, a heavy agglomeration of IFR particles is formed, and there is a clear interface between the IFR and PP matrix, confirming the incompatibility between the IFR and PP matrix. However, the cross sections become smooth and the dispersion of IFR in the PP matrix is significantly improved for PP6 and PP7. The IFR particles get smaller and are almost encapsulated in the PP matrix. Especially for PP7, there is a better uniform dispersion of IFR particles in the PP matrix. These results confirm that PDPFDE can effectively promote the compatibility of IFR in the PP matrix. Moreover, the molecular structure of PDPFDE contains ethylenediamine soft segment structure, which can buffer the internal stress of PP composite and play a toughening effect.50,51

The tensile strength and stress‐strain curves of PP and its composites are shown in Figure 8A,B, and the corresponding data are listed in Table S1. It can be seen that the tensile strength of PP0 is 38.13 MPa, and its elongation at break is 46.27%. Because of the poor dispersion of IFR in the PP matrix, both tensile strength and elongation at break of PP1 and PP5 are reduced, which makes the PP composites appear a tendency to become brittle. With the loading of PDPFDE, although the tensile strength of PP6 and PP7 are not improved, the elongation at break of PP6 and PP7 is increased by 99.0% and 247.6%, respectively, compared with PP5. It shows that PDPFDE can greatly improve the toughness of PP composites by strengthening the compatibility of IFR in PP (Figure 7). This is attributed to the fact that PDPFDE not only has good compatibility with the polymer matrix due to its organic structural units (CH2CH2, benzene ring, and Cp ring)27 but also has a strong affinity with APP because both PDPFDE and APP have the same NH groups. Therefore, PDPFDE can promote the uniform dispersion of IFR in the PP matrix like a coupling agent. In short, PDPFDE can act as a plasticizer in the PP matrix, which is beneficial for the process properties of PP/IFR composite.


By means of the synergistic flame retardant effect between PDPFDE and IFR, the flame retardant efficiency of IFR in PP is improved and the total loading amount of IFR is reduced. A suitable amount of PDPFDE can increase the thermal stability and the char residue of the PP/IFR composites. Adding 0.5 wt% PDPFDE and 22.5 wt% IFR, the LOI value of PP6 reaches 28.5%, an UL‐94 V‐0 rating is achieved, and the pHRR and THR values of PP reduced by 63% and 43%, respectively. Raman spectroscopy reveals that PDPFDE can promote the formation of graphitized char residue in PP/IFR/PDPFDE composites. Simultaneously, the loading of PDPFDE can significantly improve the dispersion of IFR in PP matrix, and the impact strength and tensile elongation at break of PP/IFR composites are visibly increased owing to the plasticizing effect of PDPFDE.


This paper was financially supported by the National Natural Science Foundation of China (51373140 and 51673160), The Project of State Key Laboratory of Environment‐Friendly Energy Materials, Southwest

University of Science and Technology (17FKSY0116), Longshan Academic Talent Research Support Plan of Southwest University of Science and Technology (17LZX404 and 18LZX440), and Postgraduate Innovation Fund Project by Southwest University of Science and Technology (19ycx0009).


Xiao‐Ping Hu


  1. Chen HD, Wang JH, Ni AQ, Ding AX, Sun ZH, Han X. Effect of novelintumescent flame retardant on mechanical and flame retardant properties of continuous glass fibre reinforced polypropylene composites.

Compos Struct. 2018;203:894‐902.

  1. Wang PJ, Hu XP, Liao DJ, et al. Dual fire retardant action: the combined gas and condensed phase effects of azo‐modified NiZnAl layered double hydroxide on intumescent polypropylene. Ind Eng Chem Res. 2017;56(4):920‐932.
  2. Xu B, Wu X, Ma W, Qian LJ, Xin F, Qiu Y. Synthesis and characterization of a novel organic‐inorganic hybrid char‐forming agent and its flame‐retardant application in polypropylene composites. J Anal Appl Pyrol. 2018;134:231‐242.
  3. Wang PJ, Liao DJ, Hu XP, et al. Facile fabrication of biobased PNCcontaining nano‐layered hybrid: preparation, growth mechanism and its efficient fire retardancy in epoxy. Polym Degrad Stab.


  1. Crewe RJ, Andrews SP, Hull TR, Birbeck PN, Stec AA. Combustion andtoxic gas production from disposable barbecues in enclosures. J Forensic Sci. 2014;59(1):127‐138.
  2. Yang S, Wang J, Huo SQ, Cheng LF, Wang M. The synergistic effect ofmaleimide and phosphaphenanthrene groups on a reactive flameretarded epoxy resin system. Polym Degrad Stab. 2015;115:63‐69.
  3. DiGangi J, Blum A, Bergman Å, et al. San Antonio statement on brominated and chlorinated flame retardants. Environ Health Perspect.


  1. Velencoso MM, Battig A, Markwart JC, Schartel B, Wurm FR. Molecular firefighting‐how modern phosphorus chemistry can help solve the challenge of flame retardancy. Angew Chem Int Ed.


  1. Pappalardo S, Russo P, Acierno D, Rabe S, Schartel B. The synergisticeffect of organically modified sepiolite in intumescent flame retardant polypropylene. Eur Polym J. 2016;76:196‐207.
  2. Wu ZP, Hu YC, Shu WY. Effect of ultrafine zinc borate on the smokesuppression and toxicity reduction of a low‐density polyethylene/intumescent flame‐retardant system. J Appl Polym Sci.


  1. Atabek Savas L, Mutlu A, Dike AS, Tayfun U, Dogan M. Effect of carbon fiber amount and length on flame retardant and mechanical properties of intumescent polypropylene composites. J Compos Mater.


  1. Lu LG, Guo N, Qian XD, et al. Thermal degradation and combustionbehavior of intumescent flame‐retardant polypropylene with novel phosphorus‐based flame retardants. J Appl Polym Sci.


  1. Huang GB, Wang SQ, Song PA, Wu CL, Chen SQ, Wang X. Combination effect of carbon nanotubes with graphene on intumescent flame‐retardant polypropylene nanocomposites. Compos Part A‐Appl S. 2014;59:18‐25.
  2. Song PA, Xu LH, Guo ZH, Zhang Y, Fang ZP. Flame‐retardant‐wrapped carbon nanotubes for simultaneously improving the flame retardancy and mechanical properties of polypropylene. J Mater Chem.


  1. Wen PY, Wang D, Liu JJ, Zhan J, Hu Y, Yuen KK. R. Organically modified montmorillonite as a synergist for intumescent flame retardant against the flammable polypropylene. Polym Adv Technol.


  1. Wang ZJ, Liu YF, Li J. Regulating effects of nitrogenous bases on thechar structure and flame retardancy of polypropylene/intumescent flame retardant composites. ACS Sustain Chem Eng.


  1. Chen HD, Wang JH, Ni AQ, Ding AX, Han X, Sun ZH. The effects of amacromolecular charring agent with gas phase and condense phase synergistic flame retardant capability on the properties of PP/IFR composites. Materials. 2018;11(1):111.
  2. Cao K, Wu SL, Qiu SL, Li Y, Yao Z. Synthesis of N‐alkoxy hindered amine containing silane as a multifunctional flame retardant synergist and its application in intumescent flame retardant polypropylene. Ind Eng Chem Res. 2012;52:309‐317.
  3. Iqbal N, Khan MS, Adnan Saeed M, Akhter Z. Synthesis and physicochemical studies of ferrocene‐containing materials. Appl Organomet Chem. 2007;21(10):862‐869.
  4. Li Z, Expósito DF, González AJ, Wang DY. Insightful investigation ofsmoke suppression behavior and mechanism of polystyrene with ferrocene: an important role of intermediate smoke. Fire Mater.


  1. Zhou KQ, Zhang QJ, Liu JJ, et al. Synergetic effect of ferrocene andMoS2 in polystyrene composites with enhanced thermal stability, flame retardant and smoke suppression properties. RSC Adv.


  1. Wen Y, Cheng Z, Li WX, et al. A novel oligomer containing DOPO andferrocene groups: synthesis, characterization, and its application in fire retardant epoxy resin. Polym Degrad Stab. 2018;156:111‐124.
  2. Kishore K, Kannan P, Iyanar K. Synthesis, characterization, and fireretardancy of ferrocene containing polyphosphate esters. J Polym Sci Pol Chem. 1991;29(7):1039‐1044.
  3. Chen PF, Yan KK, Leng JX, Zhang F, Jiao L. Synergistic smoke suppression effect of epoxy cross‐linked structure and ferrocene on epoxybased intumescent flame‐retardant coating. Plast Rubber Compos.


  1. Zhou KQ, Gui Z, Hu Y. The influence of graphene based smoke suppression agents on reduced fire hazards of polystyrene composites.

Compos Part A‐Appl S. 2016;80:217‐227.

  1. Mehdipour‐Ataei S, Babanzadeh S. Synthesis, characterization and properties of novel polyamides containing ferrocene unit and flexible spacers. Appl Organomet Chem. 2007;21(5):360‐367.
  2. Liao DJ, Xu QK, McCabe RW, et al. Ferrocene‐based non‐phosphorus copolymer: synthesis, high‐charring mechanism and its application in fire retardant epoxy resin. Ind Eng Chem Res.


  1. Liu L, Chen XL, Jiao CM. Influence of ferrocene on smoke suppressionproperties and combustion behavior of intumescent flame‐retardant epoxy composites. J Therm Anal Calorim. 2015;122(1):437‐447.
  2. Nisar M, Bergmann CP, Geshev J, Quijada R, Galland GB. Synthesis andcharacterization of polypropylene/iron encapsulated carbon nanotube composites with high magnetic response at room temperature. Polymer. 2017;118:68‐74.
  3. Nisar M, Bergmann C, Geshev J, Quijada R, Galland GB. An efficientapproach to the preparation of polyethylene magnetic nanocomposites. Polymer. 2016;97:131‐137.
  4. Lewin M, Endo M. Catalysis of intumescent flame retardancy of polypropylene by metallic compounds. Polym Adv Technol. 2003;14(1):3‐11.
  5. Zhang Y, Li XN, Fang ZP, Hull TR, Kelarakis A, Stec AA. Mechanism ofenhancement of intumescent fire retardancy by metal acetates in polypropylene. Polym Degrad Stab. 2017;136:139‐145.
  6. Brockwell T, Metcalfe E. Pyrolysis‐mass spectral studies on ferrocenyl compounds: evidence for the norbiphenylene cation. Tetrahedron.


  1. Schartel B, Hull TR. Development of fire‐retarded materials—interpretation of cone calorimeter data. Fire Mater. 2007;31(5):327‐354.
  2. Wu ZZ, Ni YP, Fu T, et al. Effect of biphenyl biimide structure on thethermal stability, flame retardancy and pyrolysis behavior of PET.

Polym Degrad Stab. 2018;155:162‐172.

  1. Chen YJ, Wang W, Qiu Y, Li LS, Qian LJ, Xin F. Terminal group effectsof phosphazene‐triazine bi‐group flame retardant additives in flame retardant polylactic acid composites. Polym Degrad Stab.


  1. Shen YL, Gong WG, Zheng BC, Meng X, Gao L. Synergistic effect ofNi‐based bimetallic catalyst with intumescent flame retardant on flame retardancy and thermal stability of polypropylene. Polym Degrad Stab.


  1. Yuan BH, Fan A, Yang M, et al. The effects of graphene on the flammability and fire behavior of intumescent flame retardant polypropylene composites at different flame scenarios. Polym Degrad Stab.


  1. Feng CM, Liang MY, Jiang JL, Huang JG, Liu HB. Synergistic effect of anovel triazine charring agent and ammonium polyphosphate on the flame retardant properties of halogen‐free flame retardant polypropylene composites. Thermochim Acta. 2016;627‐629:83‐90.
  2. Feng CM, Liang MY, Huang JG, Liu HB. Flame retardant properties andmechanism of intumescent flame retardant polypropylene/La2O3 composites. Polym Mater Sci Eng. 2013;29(6):92‐95.
  3. Yang AH, Deng C, Chen H, Wei YX, Wang YZ. A novel Schiff‐base polyphosphate ester: highly‐efficient flame retardant for polyurethane elastomer. Polym Degrad Stab. 2017;144:70‐82.
  4. Qin ZL, Li DH, Li Q, Yang RJ. Effect of nano‐aluminum hydroxide on mechanical properties, flame retardancy and combustion behavior of intumescent flame retarded polypropylene. Mater Design.


  1. Wang JJ, Wang L, Xiao AG. Recent research progress on the flameretardant mechanism of halogen‐free flame retardant polypropylene.

Polym‐Plast Technol. 2009;48(3):297‐302.

  1. Zhang P, Hu Y, Song L, Lu HD, Wang J, Liu QQ. Synergistic effect ofiron and intumescent flame retardant on shape‐stabilized phase change material. Thermochim Acta. 2009;487(1):74‐79.
  2. Montoya MI, Planas E, Casal J. A comparative analysis of mathematicalmodels for relating indoor and outdoor toxic gas concentrations in accidental releases. J Loss Prevent Proc. 2009;22(4):381‐391.
  3. Nie SB, Liu L, Dai GL, Zhou C. Investigation on pyrolysis of intumescent flame‐retardant polypropylene (PP) composites based on synchrotron vacuum ultraviolet photoionization combined with molecular‐beam mass spectrometry. J Therm Anal Calorim.


  1. Magrez A, Smajda R, Seo JW, et al. Striking influence of the catalystsupport and its acid‐base properties: new insight into the growth mechanism of carbon nanotubes. ACS Nano. 2011;5(5):3428‐3437.
  2. Kong QH, WuT, Wang JQ, Liu H, Zhang JH. Improving the thermal stability and flame retardancy of PP/IFR composites by NiAl‐layered double hydroxide. J Nanosci Nanotechnol. 2018;18(5):3660‐3665.
  3. Lai XJ, Zeng XR, Li HQ, Liao F, Zhang HL, Yin CY. Preparation andproperties of flame retardant polypropylene with an intumescent system encapsulated by thermoplastic polyurethane. J Macromol Sci, Part B. 2012;51(1):35‐47.
  4. Shao ZB, Deng C, Tan Y, et al. Ammonium polyphosphate chemicallymodified with ethanolamine as an efficient intumescent flame retardant for polypropylene. J Mater Chem A. 2014;2(34):13955‐13965.
  5. Lan YH, Li DH, Yang RJ, Liang WS, Zhou LX, Chen ZW. Computer simulation study on the compatibility of cyclotriphosphazene containing aminopropylsilicone functional group in flame retarded polypropylene/ammonium polyphosphate composites. Compos Sci Technol. 2013;88:9‐15.


Additional supporting information may be found online in the Supporting Information section at the end of the article.

How to cite this article: Li W‐X, Liao D‐J, Hu X‐P, Cheng Z, Xie C‐Q. Synergistic improvement of fire retardancy and mechanical properties of ferrocene‐based polymer in intumescent polypropylene composite. Polym Adv Technol.


word image 715


A novel oligomer containing DOPO and ferrocene groups: Synthesis, characterization, and its application in fire retardant epoxy resin

Wen, Yi, Cheng, Zhou, Li, Wenxiong, Liao, Duijun, Huang, Xiaoping, Pan, Ning, Wang, Deyi and Hull, T Richard

Available at

Wen, Yi, Cheng, Zhou, Li, Wenxiong, Liao, Duijun, Huang, Xiaoping, Pan, Ning, Wang, Deyi and Hull, T Richard ORCID: 0000­0002­7970­4208 (2018) A novel oligomer containing DOPO and ferrocene groups: Synthesis, characterization, and its application in fire retardant epoxy resin. Polymer Degradation and Stability, 156 . pp. 111­124. ISSN 0141­3910

It is advisable to refer to the publisher’s version if you intend to cite from the work.

For more information about UCLan’s research in this area go to and search for <name of research Group>.

For information about Research generally at UCLan please go to

All outputs in CLoK are protected by Intellectual Property Rights law, including Copyright law. Copyright, IPR and Moral Rights for the works on this site are retained by the individual authors and/or other copyright owners. Terms and conditions for use of this material are defined in the

word image 716 CLoK

Central Lancashire online Knowledge

Publication data: Wen, Y., et al, (2018) Polymer Degradation and Stability, 156, 111-124.


A novel oligomer containing DOPO and ferrocene groups:

synthesis, characterization, and its application in fire retardant epoxy resin

Yi Wen a, Zhou Cheng a, Wenxiong Li a, Zhi Lic, Duijun Liao a, Xiaoping Hu a*, Ning Pan b, Deyi Wangc, T. Richard Hull d*

aState Key Laboratory for Environment-friendly Energy Materials, School of Materials Science and Engineering, Southwest University of Science and Technology, Mianyang 621010, P. R. China b Fundamental Science on Nuclear Wastes and Environmental Safety Laboratory, Southwest University of Science and Technology, Mianyang 621010, P. R. China cIMDEA Materials Institute, C/Eric Kandel, 2, 28906 Getafe, Madrid, Spain d Centre for Fire and Hazards Science, University of Central Lancashire, Preston PR1 2HE, UK


*Corresponding Author. E-mail address: (Xiaoping Hu), (T. Richard Hull)



A novel oligomer (PFDCHQ) based on 9,10-dihydro-9-oxa-10-phosphaphenanthrene -10-oxide (DOPO) and ferrocene groups was synthesized successfully, aiming at improving the flame retardant efficiency of diglycidyl ether of bisphenol A epoxy resin (DGEBA). FTIR, 1H NMR and 31P NMR were used to confirm the chemical structure of PFDCHQ. The high char yields of 60.3 wt% and 20.1 wt% were obtained for PFDCHQ from TGA results in nitrogen and air atmosphere, respectively. The thermal degradation mechanism of PFDCHQ was investigated by TG-FTIR and Py-GC/MS. The limiting oxygen index (LOI) of EP-5 with 5 wt% loading of PFDCHQ increased to 32.0% and the UL-94 V-0 rating was achieved, showing a notable blowing-out effect . In contrast to EP-0, the peak of the heat release rate (pHRR) and total heat release (THR) of EP-5 decreased by 18.0% and 10.3%. The flame retardant mechanism of PFDCHQ in epoxy resin was studied by TG-FTIR, SEM and Raman. SEM and Raman results indicated the formation of coherent and dense char residue with high degree of graphitization due to the incorporation of PFDCHQ. In UL-94, the blowing-out effect dominantly accounted for the enhanced flame retardancy in combination with optimized char structure. Furthermore, the addition of PFDCHQ improved the Young’s modulus compared to EP-0.

Keywords: DOPO-HQ; Ferrocene; Synthesis; Epoxy resin; Flame retardant mechanism



Epoxy resin (EP) is widely applied in many areas such as coating, adhesive, laminating, electronic/electrical insulation, and composite application [1-3], due to its excellent mechanical properties, chemical stability, bonding behavior and abundant product forms [4-6]. However, the high flammability is a general shortcoming of EP and limits its applications. Therefore, it has attracted more and more attention to endow EP with required flame-retardance.


As an important halogen-free flame retardant, phosphorus flame retardants (PFRs) [7, 8] have attracted extensive attention [9-11] for their high efficiency. Most of the PFRs not only play flame retardant effects through a protective char layer caused by the catalyzation of phosphoric acid [12, 13], but also play flame retardant role in the gas phase through the radical trapping by PO and PO2 fragments [14, 15]. In recent years, 9,10-dihydro-9-oxa-10-phosphaphenanthrene-10oxide (DOPO) and its derivatives, as a novel kind of PFRs, have received considerable attention attributed to their high reactivity [16]. When DOPO-based PFRs are applied in epoxy resin, it is often used in conjunction with nitrogen/silicon elements as small molecular PFRs [17-19]. However, the non-polymeric PFRs display many drawbacks such as poor compatibility with polymer matrix and easily leaching. Comparatively, DOPO-containing polymeric flame retardants not only have higher phosphorus content and richer aromatic group structures, but also can overcome these shortcomings of small molecular PFRs [20, 21].


The 10-(2,5-Dihydroxyphenyl)-10H-9-oxa-10-phospha-phenantbrene-10-oxide (DOPO-HQ) is one of the most important derivatives of DOPO. The two hydroxyl groups in DOPO-HQ molecule can react with other active function groups to form polymeric phosphorus-containing flame retardants, which have good compatibility with epoxy resin and are not easily migrated during the processing or using according to the published literature [22, 23]. Wang et al. synthesized a polymeric PFR based on 10-(2, 5-dihydroxyl-phenyl)-9, 10-dihydro-9-oxa-10phosphaphenanthrene-10-oxide (DOPO-BQ) and pentaerythritol diphosphonate dichloride (SPDPC), a V-0 rating in the UL-94 test was obtained with PFR content of 10 wt% in epoxy resin [24]. Carja et al. reported a new phosphorus flame retardant by solution polycondensation of 1,4phenylenebis phenylenebis ((6-oxido-6H-dibenz[c,e][1,2]oxaphosphorinyl) carbinol) with phenylphosphonic dichloride, and a UL-94 V-0 rating material was obtained when 14.8 wt% PFR were added into the epoxy matrix [23]. In addition, Tian et al. synthesized a novel organophosphorus named as poly(4,4-dihydroxy-1-methyl-ethyl diphenol-o-bicyclic pentaerythritol phosphatephosphate) (PCPBO) and achieved good flame retardant epoxy resin composites with the formation of an intumescent char layer [25]. However, the high loadings was needed to reach the V-0 level due to the poor catalytic effect of phosphoric acid and the low flame-retardant efficiency when these reported polymeric PFRs were used in epoxy materials. Therefore, it is a preferable idea to introduce a functional group with high-effective catalyzing function onto PFR molecules for more char formation. It is already known that transition metals including Fe, Co and Ni can cause catalytic char-formation of polymer [26-28]. So, it is a promising way to combining iron-containing unit with phosphorus-containing unit.


Ferrocene has received considerable attention due to its effective smoke suppression and catalyzing crosslinking performance [29, 30]. The previous reports revealed that some polymers containing ferrocene and phosphorus had high char yield and thermal stability [31, 32]. Kishore et al. synthesized a series of polyphosphate esters containing ferrocene. Theses polyesterphosphate esters (VI, VII, VIII, IX and X) provided high char residues of 28%, 38%, 27%, 23% and 32% at 700 oC [33]. Mehdipour-Ataei reported a battery of ferrocene-based polyamides and had high char yield from 43% to 72% at 600 oC [34]. Besides, Liao synthesized a novel ferrocene-based copolymer with 62 % char residue at 700 oC under nitrogen. When applied it in epoxy resin, a UL94 V-1 rating and LOI value of 29.2% are obtained, which is attributed to the excellent catalytic charring capacity of ferrocene-based copolymer [35].


Combined the merits of DOPO and ferrocene groups, a novel oligomeric flame retardant poly 10(2,5-dihydroxyphenyl)-9,10-dihydro-9-oxa-10-phospha Phenanthrene-10-oxide-1,1’-ferrocene dimethyl ester (PFDCHQ) containing DOPO-HQ and ferrocene was synthesized and incorporated into diglycidyl ether of bisphenol A epoxy resin (DGEBA) to improve its fire retardancy. The chemical structure of PFDCHQ was characterized by FTIR, 1H NMR and 31P NMR. The fire retardancy, flame retardant mechanism and mechanical properties of EP/PFDCHQ composites were comprehensively studied. The results show that PFDCHQ is a kind of high-effective flame retardant for epoxy resin material.




Ferrocene (98%), oxalyl dichloride (98.0%) and methylene dichloride (CH2Cl2, 99.5%) were purchased from Sinpharm Chemical Reagent Co. Ltd. (China). DOPO-HQ (Huizhou sunstar technology Co. Ltd., China) was used without further purification. Sodium hypochlorite solution (active chlorine is 10%) was provided from Shanghai Aladdin Biochemical Technology Co. Ltd.

(China). Petroleum ether was provided by Tianjin Zhiyuan Chemical Reagent Co. Ltd. (China).

Epoxy resin (DGEBA, commercial name: E-44, with an epoxy value of 0.41-0.48) was supplied by

Nantong Xingchen Synthetic Material Co. Ltd. (Jiangsu China). The curing agent mPhenylenediamine (m-PDA) was purchased from Tianjin Guangfu Co. Ltd. (China).


Preparation of PFDCHQ

2.2.1. Synthesis of 1,1’-diacetylferrocene (DAF)

In a 500 ml three-necked flask with a magnetic stirrer, flux condenser and nitrogen inlet, AlCl3 (46.7 g, 0.35 mol) and 125 ml CH2Cl2 were introduced into it. The mixture was stirred at room temperature for 5 min. Then, the AlCl3 was dissolved after adding 107 ml acetyl chloride dropwise through dropping funnel. Next, ferrocene (18.6 g, 0.1 mol, dissolved in 100 ml absolute CH2Cl2) was added dropwise. The reaction mixture was stirred at room temperature for 2 h and the dark purple solution was produced. Afterwards, the solution was slowly poured into lots of ice. The organic phase and water phase were separated with a separating funnel. After that, the organic phase was washed with water and extracted, repeatedly three times, then the CH2Cl2 was rotary evaporated. The crude product was purified by petroleum ether until the washing liquid was colorless. Finally, recrystallization from water gave 20.25 g (yield: 75%) of red acicular crystal, which was named as DAF. FT-IR (KBr, cm-1) ν: 3435 (H2O); 3094 (Cp-H); 1660 (-C=O); 2980, 1456 and 1375 (-CH3); 1297, 1150, 842, 542 and 502 (Cp ring). 1H NMR (600 MHz, CDCl3, δ, ppm): 4.51 (4H, s, Cp), 4.77 (4H, s, Cp), 2.35 (6H, s, -CH3). The corresponding spectra are shown in Fig. S1.


2.2.2. Synthesis of 1,1’-Ferrocenedicarboxylic acid (FDC)

DAF (13.3 g) was added to 350 ml of 10% sodium hypochlorite solution at 65 °C and rapid stirred in the dark. A further 200 ml of sodium hypochlorite solution was added after 2 h, and kept rapidly stirring for 6 h at 65 °C. Then the reaction solution was filtered hot and acidified to pH 1-2 with concentrated hydrochloric acid, forming a copious orange precipitation. The crude product was filtered, next dissolved in sodium hydroxide solution, and subsequently, reacidified with concentrated hydrochloric acid and precipitated to obtain pure orange 1,1’-ferrocenedicarboxylic acid, which was named as FDC (10.8 g, yield: 81 %). FTIR (KBr, cm-1) ν: 3435 (H2O), 3200-2500 (association of -OH), 1688,1302 (COO) [36]; 1492, 1169 and 841 (Cp ring); 1405, 919 (O-H); 1H NMR (600 MHz, CH3COOH-d4, δ, ppm): 4.57 (4H, s, Cp), 4.93 (4H, s, Cp), 11.59 (-COOH). The FTIR and 1H NMR spectra are shown in Fig. S2.


2.2.3. Synthesis of PFDCHQ

1,1’-Ferrocenedicarbonly chloride was prepared according to previous work [37]. In a 500 ml three-necked flask equipped with a magnetic stirrer and condenser, FDC (13.0 g), CH2Cl2 (200 ml), oxalyl chloride (20 ml) and 10 drops of pyridine were added and the reaction mixture was stirred at room temperature for 12 h. Subsequently, the mixture was heated to reflux temperature for 6 h. Then, the solvent and unreaction oxalyl chloride were removed by rotary evaporated under reduced pressure to obtain crude product. Finally, the crud product was extracted repeatedly with hot petroleum until the petroleum is colorless and the crimson product was obtained, which is named as 1,1’-Ferrocenedicarbonly chloride.


DOPO-HQ (0.03 mol ) was dispersed in 150 ml absolute CH2Cl2 and added into a 500 ml threenecked round-bottom flask equipped a magnetic stirrer, condenser and constant pressure dropping funnel. Next, Et3N (8.3 ml) was introduced into the mixture and stirred at room temperature for 10 min. Thereafter, 1,1’-ferrocenedicarbonly chloride (0.03 mol), which was dissolved in 50 ml CH2Cl2, was added dropwise. Moreover, the reaction mixture was stirred at room temperature for 12 h under the nitrogen protection. Then, the brown solution was poured into 500 ml methanol with stirring after filtered and washed thoroughly with methanol, the resulting product was dried at 60 oC under vacuum to a consent mass (yield: 85 %), which is named as PFDCHQ. The preparation route of PFDCHQ is shown in Scheme1.



Scheme 1. Synthetic route to PFDCHQ

word image 2751


Preparation of EP/PFDCHQ composites

Briefly, the preparation of epoxy composites with 5 wt% PFDCHQ was as follows: epoxy resin (50.0 g) and PFDCHQ (2.63 g) were added into a 150 ml flask with 20 ml dry CH2Cl2 and stirred 10 min. After a homogeneous mixture was obtained, and then placed to the rotary evaporators at 70 oC to remove the CH2Cl2 solvent. Next, m-PDA (5.0 g) was added in the mass ratio 10: 1 of EP.

After that, the mixture was stirred at 80 oC under reduced pressure until no bubbles emerged. Subsequently, it was poured into a preheated standard polytetrafluoroethylene (PTFE) mold at 80 oC . The mixture was cured at 80 oC for 3 h, 100 oC for 2 h and then 120 oC for 3 h. After curing, the samples were cooled to room temperature and the EP/PFDCHQ was obtained. The formulas of EP/PFDCHQ composites are listed in Table 1. Other samples were prepared in the same procedure. A schematic representation of the preparation process of EP/PFDCHQ composites is shown in Scheme 2.



word image 2752

Scheme 2. A schematic representation of the preparation process of EP/PFDCHQ composites



Fourier Translation Infrared Spectroscopy (FTIR) of PFDCHQ and its intermediate on a KBr pellet were analyzed on a FTIR spectrometer (Nicolet 5700) over the wavenumber from 400 to 4000 cm-1.

1H NMR and 31P NMR spectra were performed on Bruker Avance 600 spectrometer (600 MHz), using DMSO-d6 as solvent.

The glass transition temperature (Tg) of PFDCHQ with ~5 mg was measured by differential scanning calorimeter (DSC) on Jupiter STA 449C thermal analyzer (Netzsch, Germany) with a heating rate of 10 oC/min ranging from 25 oC to 200 oC under nitrogen atmosphere.


Table 1. Formulation of EP/PFDCHQ composites

































Samples EP (g) m-PDA (g)


The DSCQ2000 (TA Instrument company, USA) was used to study Tg of EP and its composite under nitrogen. All samples were heated from room temperature to 150 oC at a heating rate 10 oC/min and keeping for 5 min, and then cooled to room temperature at -20 oC/min. A second scanning was conducted at the same heating rate as the first time from room temperature to 150 oC.


The molecular weight and their distribution were determined by Gel Permeation

Chromatography (GPC) with an Agilent 1200SERIES instrument using THF as eluent at a flow rate of 1.000 ml/min.

The thermogravimetric analysis (TGA) of samples with ~5 mg was carried out with STA6000 simultaneous thermal analyzer (PerkinElmer, USA) from 40 oC to 700 oC at a heating rate of 10 oC/min under nitrogen.


The pyrolysis behavior (Py-GC/MS) of the PFDCHQ was tested with DANI MASTER GC-TOF-MS system combined with a pyrolyzer (CDS5200). The samples (~0.3 mg), packed in a quartz tube capillary using the platinum coil attachment, were heated from ambient temperature to 700 oC at a rate of 1000 oC/s and kept at this state for 20 seconds. The Py/GC interface temperature was set at 100 oC. The transfer line temperature was set at 260 oC. The injector temperature was set at 280 oC and operated in the split mode (split ratio 1000: 1) with helium as carrier gas. The detailed data were analyzed using the NIST Mass Spectral Search Program and the NIST library was employed as the standard spectral library to match the volatile pyrolysis product recorded from the analysis.

Thermogravimetric analysis/fourier transform infrared spectra (TG-FTIR) was performed on a STA6000 simultaneous thermal analyzer (PerkinElmer, USA) at a heating rate of 10 oC/min from 40 oC to 700 oC with ~20 mg powder samples under nitrogen atmosphere.


The limiting oxygen index (LOI) values of the samples were measured by a JF-3 oxygen index meter (Nanjing Jionglei Instrument Equipment Co., Ltd) according to the standard of ASTM D 2863-97 with the three dimensions size of 100.0×6.5×3.2 mm3.


UL-94 vertical burning tests were performed on a vertical burning test instrument (Nanjing Jionglei Instrument Equipment Co., Ltd) based on the standard of ASTM D 3801 with the three dimensions size of 130.0×13.0×3.2 mm3.


The cone calorimeter test was carried out on a FTT Standard Cone Calorimeter (Fire Testing Technology, UK) according to ISO 5660 under an external heat flux of 35 kW/m2 with sample dimensions of 100.0×100.0×3.0 mm3.


The char residues of EP and EP/PFDCHQ composites after the UL-94 test were investigated on a TM-3000 (Hitachi, Japan) desktop scanning electron microscope (SEM).


The PIT501J LCD plastic Charpy impact testing machine (Shenzhen million Test Equipment Co., Ltd.) was utilized to measure the impact performance of pure EP and EP/PFDCHQ composites with their dimensions of 80.0×10.0×4.0 mm3, according to GB/T1043-2008. The tensile strength was performed on ETM105D 100 KN computer-controlled electronic universal testing machine (WANCE GROUP) with a testing speed of 20 mm/min, depending on GB/T1040.3-2006.


The fracture surface of EP/PFDCHQ composites was carried on the Ultra55 scanning electron microscope (SEM).


Results and discussion

Characterization of PFDCHQ

The structure of PFDCHQ was confirmed by FTIR, 1H NMR and 31P NMR. Some specific absorption peaks appear from DOPO-HQ (Fig.1(a)): 1595 cm-1 (P-Ph), 1197 cm-1 (P=O), 924 cm-1 (P-O-Ph) [38]. In case of PFDCHQ, these characteristic peaks still exist. The peaks at 3107, 1583 cm-1 (benzene ring) and 1450, 832, 493 cm-1 (Cp ring) [39] are clearly observed in the FTIR spectrum of PFDCHQ. In addition, the most significant absorptions are observed at 1737 cm-1 and 1092 cm-1, 1181 cm-1, 1274 cm-1 corresponding to the C=O and C-O-C respectively [12], indicating the existence of ester group. In the 1H NMR spectrum of PFDCHQ (see Fig.1 (b)), the peaks ranging from 8.09 to 7.10 ppm are ascribed to the protons on the benzene ring. The peaks from 5.50 to 4.01 ppm belong to the protons on the Cp ring. Furthermore, PFDCHQ exhibits a single peak at 33.9 ppm in 31P NMR spectrum (Fig. 1(c)). All the results of FTIR, 1H NMR and 31P NMR state that PFDCHQ is obtained successfully.


word image 2753

Fig. 1. (a) FTIR spectra of DOPO-HQ and PFDCHQ, (b) 1H NMR spectrum and (c) 31P NMR spectrum of PFDCHQ


The GPC and DSC curves of PFDCHQ are showed in Fig. S3. The weight average molecular weight

(Mw) of PFDCHQ is 10030 g/mol and a number molecular weight (Mn) is 6082 g/mol, accompanied by the polydispersity Mw/Mn of 1.649. The glass transition temperature (Tg) value of

92.4 oC suggests a relatively high temperature resistance of PFDCHQ.


Thermal and pyrolysis behavior of PFDCHQ

The TGA/DTG curves of PFDCHQ in both nitrogen and air are shown in Fig. 2. Based on Fig. 2(a), there are two thermal degradation stages. The first one ranges from 230.6 to 326.6 oC, and a weight loss of 5 wt% occurs due to the thermal cleavage of ester bond and the early thermal decomposition of some PFDCHQ with low-molecular weight, generating DOPO and ferrocenebased segments. The main degradation stage takes place from 342.4 to 597.5 oC, accompanied by the temperature of maximum mass loss rate (Tmax) around 481.2 oC and a weight loss of 35 wt%. At this stage, the DOPO-based segments are mainly decomposed into dibenzofuran, PO and PO2 radicals in nitrogen environment due to the lack of oxygen. According to the TG-FTIR data (Fig. 3), partial ferrocene-containing segments begin to decompose above 390 oC. As for the thermal degradation of PFDCHQ in air (Fig. 2 (b)), the more complex thermal-oxide degradation process can be seen. The initial decomposition temperature (T-5%) of PFDCHQ is advanced to 269.1 oC due to the existence of oxygen. The first main decomposition stage from 218.9 oC to 523.4 oC with a weight loss of 38.5%, including series of thermal degradation stages from DTG curve, can be assigned to the decomposition of low-molecule weight PFDCHQ, cleavage of ester bonds, and the further degradation of ferrocene or phosphorus-containing structures, generating an unstable carbonaceous layer. The second main thermal degradation stage from 523.4-622.5 oC originates from the further thermal-oxide decomposition of the unstable carbonaceous layer. Moreover, the char residues of PFDCHQ in nitrogen and air atmosphere are 60.3 wt% and 20.1 wt% at 700 oC, respectively, indicating that PFDCHQ has better thermal stability during high-temperature stage in nitrogen atmosphere.

word image 2754

Fig. 2. TG and DTG curves of PFDCHQ in nitrogen (a) and air (b) atmosphere


TG-FTIR was usually used to analyze the gaseous products during the thermal degradation. The FTIR spectra of pyrolysis products of PFDCHQ at different temperatures are shown in Fig. 3. The relevant characteristic absorbing peaks appear at 3738 cm-1 (H2O), 3015 cm-1 (Cp ring), 2360 cm-1 (CO2) and 1508 cm-1 (aromatic compounds and Cp ring) [40, 41]. Before 250 oC, there is no obvious peak of CO2 indicating that PFDCHQ has a good thermal stability. With the temperature increasing to 390 oC, the new absorbing peak at 3015 cm-1 (Cp ring) comes into sight [21], which is attributed to the decomposition of ferrocene-containing compounds at high temperature. New peaks of 1262 cm-1 and 1180 cm-1 at 400 oC can be attributed to P=O and C-O-C generated from DOPO-containing segments [42, 43]. When the temperature rises to 500 oC, the absorption intensity of P=O, C-O-C and aromatic compounds are weakened, and it can be inferred that the rest of P element and aromatic compounds remain in the condensed phase and take part in the formation of char residue [44].

word image 2755

Fig. 3. The TG-FTIR spectra of volatilized products at different temperatures during thermal decomposition of PDPFDE under N2

To further identify the thermal degradation mechanism of PFDCHQ, Py-GC/MS test was performed. The total ion chromatogram and analysis data are shown in Fig. 4 and Table S1. The predominant decomposition volatiles of PFDCHQ are ethanol, phenol, 2-methyphenol, biphenylene, dibenzofuran, o-hydroxybiphenyl and fluorine, corresponding to the peaks 1, 2, 3, 6, 7, 8 and 9, respectively. Impressively, some aromatic compounds are produced, such as naphthalene (peak 4), fluorine (peak 9) and phenanthrene (peak 10).

word image 2756

Fig. 4. Total ion chromatogram of PFDCHQ


The pyrolysis mechanism of PFDCHQ can be further elucidated based on the cracking product in Py-GC/MS and shown in Scheme 3. With the increasing temperature, the O=C-O bond is firstly dissociated; decomposing into two parts of ferrocene-based group and DOPO-containing species.

Subsequently, the groups on the two sides of ferrocene derivatives begin to cleave, producing CO2 and ferrocene. Besides, the derivatives containing DOPO groups begin to decompose to produce 6-methyl-6H-dibenzo[c,e][1,2]oxaphosphinine 6-oxide (peak 11 in Fig. 4). It is possible to speculate that 6-methyl-6H-dibenzo[c,e][1,2]oxaphosphinine 6-oxide could release DOPO free radical, then proceed to generate PO2 and PO radicals, accompanied by biphenylene and dibenzofurans, respectively [45, 46], since the biphenylene (peak 6) and dibenzofuran (peak 7) were detected in the chromatogram of PFDCHQ (Fig. 4). The PO2 and PO radicals can contribute to flame inhibition by trapping active H and HO free radicals and stopping the chain reaction [47]. The analysis result of Py-GC/MS is consistent with the TG-FITR and TG data of PFDCHQ under nitrogen condition.



word image 2757

Scheme. 3. Possible pyrolytic route of PFDCHQ during thermal degradation


Thermal properties of EP/PFDCHQ composites

Thermogravimetric analysis is the most common method for evaluation of the thermal decomposition property of polymeric materials. Fig. 5 illustrates the TG and DTG curves of EP-0, EP-5 and EP-6 in nitrogen atmosphere. The onset degradation temperature (Td) of samples is evaluated by the temperature at 5 wt% weight loss (T-5%) and the char residues at 700 oC are obtained from the TG curves; the temperature at maximum weight loss rate (Tmax) of samples is obtained from the DTG curves. Some important data are listed in Table 2. The DTG curves of EP-0, EP-5 and EP-6 present one decomposition step with maximum mass loss rate at 368.2, 351.5 and 357.1 oC respectively. The T-5% of EP-5 and EP-6 are lower than EP-0, which may be attributed to the fact that the introduction of ferrocene group can promote the advanced thermal degradation of epoxy resin [35], which would provide more carbon source to form protective char layer. Combining the self-high-charring properties of PFDCHQ, the resulting char residue of EP-5 (25.1 wt%) and EP-6 (22.1 wt%) are both higher than EP-0 (16.8 wt%) at 700 °C.


Table 2. TGA data of EP-0, EP-5 and EP-6


T-5% (°C)

Tmax (°C)

Char residues

(700 °C) wt%














Another important parameter for evaluating the thermal stability of epoxy resin is glass transition temperature (Tg). Thus DSC results of EP/PFDCHQ composites are shown in Fig. 6. The Tg value of EP-0 is 122 oC and the Tg value of EP-5 increases to 131 oC with 5 wt% PFDCHQ loading. This may be attributed to the rigid structure (ferrocene unit and benzene ring ) of PFDCHQ, which can limit the movement of epoxy resin molecular chain [48]. As the content of PFDCHQ reaches to 6 wt%, the Tg value of EP-6 decrease slightly due to the aggregation of excessive PFDCHQ.


word image 2758

Fig. 5. TG (a) and DTG (b) curves of EP-0, EP-5 and EP-6 in nitrogen



word image 2759

Fig. 6. DSC curves of EP-0, EP-5 and EP-6


Fire behavior and flame-retardancy

The limiting oxygen index (LOI) and UL-94 test are widely used to evaluate the flammability of materials. The LOI values and the experiment results of UL-94 are shown in Table 3. The EP-0 burns quickly after ignition in both LOI and UL-94 tests. When adding 5 wt% PFDCHQ, the LOI value of EP-5 can reach 32.0% and easily pass the V-0 rating. The real-time digital photos of some samples (EP-0, EP-3, EP-5 and EP-6) during UL-94 tests are shown in Fig. 7. After evacuating the source of fire, EP-0 continues to burn without a sign of extinguish, while EP-5 was extinguished naturally in 10 s, indicating that the incorporation of PFDCHQ do improve the fire retardancy for EP, which is attributing to the three reasons: 1) The PO and PO2 radicals generated from DOPO structure inhibit flame by trapping H and HO fragments; 2) The addition of PFDCHQ promote the initial catalytic charring action on the surface of EP-5 [49]; 3) it is notable that an obvious blowing-out effect can be observed for EP-5, which is very important to wipe the flame away rapidly in UL-94 testing [50-52]. However, the LOI values and UL-94 ratings of EP-6 and EP-7 decrease probably due to two reasons: 1) the aggregation of excessive PFDCHQ; 2) the more ferrocene groups should result in a faster degradation rate than the charring rate of EP composite during combustion. These results suggest that an appropriate loading amount of PFDCHQ is a key factor for the formation of suitable strong char layer to satisfy the demanded inner volatile gas pressure to blowing-out effect.



word image 2760

Fig. 7. The digital photos during vertical burning test of EP-0, EP-3, EP-5 and EP-6




Table 3. LOI and UL-94 data of EP and its composites




LOI (%)








24.5 ± 0.3





29.2 ± 0.2






30.6 ± 0.4






32.0 ± 0.3






30.9 ± 0.2






30.3 ± 0.3





t1 and t2 are the extinguish time after two 10 second ignition; NR is no rating.


Cone calorimeter (CC) test is widely used to detect the combustion characteristic of the polymeric materials [53]. Fig. 8 showed the heat release rate (HRR), total heat release (THR), smoke production release (SPR) and mass curves of EP-0, EP-5 and EP-6. Other various important data obtained from measurements in cone calorimeter, such as the time-to-ignition (TTI), peak heat release rate (pHRR), time-to-pHRR (t-pHRR), average effective heat of combustion (AEHC), fire growth rate (FIGRA), average specific extinction area (ASEA) and the char residue yield, are summarized in Table 4. The pHRR is one of the most important fire behavior parameters for fire retarding materials. In Fig. 8 (a, b), the pHRR of EP-5 and EP-6 reduced to 939.3 kW/m2, 872.3 kW/m2, meanwhile the THR decreased to 89.2 MJ/m2 and 83.5 MJ/m2 respectively when compared to EP-0. The reason is probably attributed to PO and PO2 free radicals, produced by DOPO structure in PFDCHQ during combustion, which can interrupt the chain reaction by capturing H and HO fragments [44, 54, 55]. Unfortunately, the results of cone calorimetry test show lower fire retardancy efficiency compared to LOI and UL-94 data. What’s more, the FIGRA of EP-5 8.2 kW/m2s is higher than that of EP-0 (6.9 kW/m2s). This may be due to the fact that the catalytic degradation effect of ferrocene is slightly stronger than the free radical quenching under continuously forced heating during cone calorimetry test. Fig. 8(c) shows the mass curves of EP-0, EP-5 and EP-6 composites. The higher char residue was obtained for EP-5, illustrating that the char layer is produced by EP-5 during burning. Furthermore, it is worth noting from the mass curves that EP-5 and EP-6 begin to decompose earlier than EP-0 due to the catalytic degradation effect of PFDCHQ. The smoke production release (SPR) of EP-0, EP-5 and EP-6 are shown in Fig. 8(d). It is found that the SPR of EP-5 decreases a lot compared to EP-0, suggesting that PFDCHQ could inhibit the production of smoke for epoxy resin to a certain extent.

word image 2761

Fig. 8. (a) HRR, (b) THR, (c) Mass loss and (d) SPR curves of EP-0, EP-5 and EP-6 from cone calorimeter tests






Table 4. Detailed CC data of EP-0, EP-5 and EP-6


TTI (s)














63± 13

170 ± 9

1146.1 ± 16.1

99.4 ± 0.4

24.71 ± 0.19

695.87 ± 9.37

10.2 ± 1.4


61 ± 4

114 ± 4

939.3 ± 59.9

89.2 ± 2.3

22.63 ± 0.48

588.67± 17.07

15.8 ± 1.0


56 ± 9

113 ± 4

872.3 ± 26.0

83.5 ± 4.1

21.69 ± 1.23

623.94 ± 21.16

14.9 ± 0.8


The CO2 and CO product rate curves of EP-0, EP-5 and EP-6 are shown in Fig. 9. From Fig. 9, it can be seen that the CO2 release rate of EP-5 and EP-6 are much lower than EP-0, while their CO production rate are higher than EP-0. The result indicates that the EP/PFDCHQ composites cannot be fully burned due to the existence of PO and PO2 radicals released by PFDCHQ, which play a distinct fire retardant role in gas phase.


word image 2762

Fig. 9. CO2 (a) and CO (b) release rate curves of EP-0, EP-5 and EP-6 from cone calorimeter tests


From Table 4, it can be found that the EP-5 and EP-6 show a slight reduction in TTI and t-pHRR, which may be assigned to the catalysis degradation effect of ferrocene group in PFDCHQ. With respect to the AEHC, when adding 5 wt%, the AEHC shows a slight decrease from 24.71 (EP-0) to 22.63 MJ/kg. The ASEA is used to measure the ability of the material to produce smoke. The ASEA value of EP-5 is 588.67 m²/kg, which is lower than that of EP-0 ( 695.87 m²/kg) indicating that PFDCHQ not only can improve the flame retardancy, but also can suppress the smoke-production of EP.


Flame retardant mechanism in both condensed and gas phases

The flame retardant mechanism of EP/PFDCHQ composites was analyzed by the condensed and gas-phase analysis. FTIR spectra of pyrolysis products of EP-0 and EP-5 composites at different temperatures were recorded during TG-FTIR test and are shown in Fig. 10. The main thermal decomposition products of EP-0 can be clearly seen, such as H2O (3600-4000 cm-1), CO 2 (23072380 cm-1), aromatic compounds (3010-3030, 1512, 825, 742 cm-1) and hydrocarbons (2800-3100 cm-1 and 1100-1250 cm-1) [56-58]. For EP-5, The new absorption bands at 1262 cm-1 and 1180 cm-1 belong to the P=O and C-O-C structures, respectively. In addition, there are no significant peaks of hydrocarbon for EP-5, indicating lower the smoke-production during combustion. It is also worth noting that the peaks of aromatic compounds (825 cm-1 and 742 cm-1) for EP-5 (Fig.10(b2)) disappeared, but the peak around 1512 cm-1 remains stronger than that of EP-0, which can be attributed to the aromatic compounds originated from PFDCHQ. This result is consistent with the TG-FTIR (Fig.3) of PFDCHQ.


word image 2763

Fig. 10. 3D FTIR and FTIR spectra of the pyrolysis products of EP-0 (a1,a2) and EP-5 (b1,b2) at different temperature


In order to further understand the change of the pyrolysis products, total and some specific products of EP-0 and EP-5 are revealed in Fig. 11. It can be seen that the total pyrolysis products for EP-5 are lower than EP-0 (Fig. 11(a)) indicated that PFDCHQ can prevent epoxy resin from further combustion. Furthermore, Fig. 11(b, c, and d) display the absorbance intensity of CO2, hydrocarbons, and aromatic compounds. Hydrocarbons and aromatic compounds belong to flammable gases and tend to aggregate into smoke particles, which increase the probability of fire accident [59]. Fortunately, the absorbance intensity of hydrocarbons and aromatic compounds from EP-5 are much lower than EP-0, implying that more of them had remained in condensed phase to form the compact protective char layer [60]. Moreover, the production of CO2 from EP-5 is higher than EP-0 and it helps to dilute flammable gases, thereby preventing further combustion of the matrix materials.


Raman spectroscopy, a powerful tool for the characterization of carbonaceous materials, has been applied to analyze the char residues of EP composites. Fig. 12 shows the Raman spectra of EP-0 and EP-5. The degree of graphic structure is evaluated by the ratio of ID/IG, where ID and IG are the integrated intensities of D (~1360 cm-1) and G (~1590 cm-1) bands [61], respectively. Generally speaking, the lower ration of ID/IG means the higher graphitization degree of char [62].

And the higher graphitized char can effectively retard the flame and protect the inner polymer matrix from further burning. The values of ID/IG follow the sequence of EP-0 (1.589)﹥EP-5 (1.137), illustrating that the addition of PFDCHQ can contribute to form a stable graphic structure char layer.

word image 2764

Fig. 11. Absorbance of pyrolysis products for EP-0 and EP-5 versus temperature: (a) gram-schmidt, (b) CO2, (c) hydrocarbons and (d) aromatic compounds


word image 2765

Fig. 12. Raman spectra of char residues of EP-0 and EP-5 composites after UL-94 tests


To further clarify the structure of the fire-resistant barrier on the surface, the char residues of EP0 and EP-5 after UL-94 tests were analyzed after UL-94 texts using FTIR and shown in Fig. 13. It is obvious that the two spectra are different between EP-0 and EP-5. The weak peak at around 1360 cm-1 of EP-5 is due to the stretching vibration of P=O [63]. Meanwhile, the new absorption peak of EP-5 appearing at 885 cm-1 is assigned to the stretching vibrations of P-O-P or P-O-Ph bonds [64, 65]. These bonds can play a role of cross-linker to link different aromatic species and strengthen carbon layers [66, 67]. The new peak at 753 cm-1 should belong to C-H deformation vibrations of aromatic ring.


word image 2766

Fig. 13. FTIR spectra of the char residues of EP-0 and EP-5 composites after UL-94 tests


The morphology of the char residues after UL-94 tests was investigated by SEM. Fig. 14(a, b, c, and d) display the morphologies of char residues from EP-0, EP-3, EP-5 and EP-6, respectively. The residue of the EP-0 shows a loose and fluffy structure. This type of carbon layer is not conducive to prevent the transmission of combustible gases and heat. The introduction of PFDCHQ in EP matrix leads to more char residues, especially for EP-5, which effectively protects the internal epoxy resin matrix when the fire contacts them. But, an important phenomenon must be emphasized: all the char residues of EP composites are not continuous and compact, which may be helpful to the abrupt release of volatile gas, generating blowing-out effect to kill flame.


Based on the analysis above, we can attempt to explain the incoherence between LOI and UL-94 and CONE tests; namely, why the pHRR value decrease slightly in contrast to the excellent LOI and UL-94 performance for EP-5. Firstly, the PO and PO2 free radicals will come into play in the early stage after ignition to inhibit the fast combustion of EP matrix, which should facilitate the formation of char layer, as can be seen in UL-94 test (Fig. 14). Secondly, a suitable compact char layer produced in UL-94 test is definitely crucial to match initial blowing-out effect for killing flame rapidly (EP-5). Less or rich char layer is not conducive to generate blowing-out effect, indicating that a proper loading of PFDCHQ is necessary and critical. However, under the continuous irradiation in the CONE test, in spite of existing radical trapping effect in gas phase, the ferrocene mainly play an catalyzing degradation role on EP matrix, resulting in no sufficient char layer to prevent further combustion of EP matrix, then a high pHRR value appears for EP-5. Actually, more investigations are needed to elucidate profoundly the fire retardancy mechanism of EP/PFDCHQ system in the future.

word image 2767

Fig. 14. Different magnifications of SEM micrographs of EP-0 (a1-a3), EP-3 (b1-b3), EP-5 (c1-c3) and EP-6 (d1-d3) after UL-94 tests


Mechanical properties

Mechanical property is the important factor to value the practical usability of composites. Impact strength, tensile strength, elongation at break and Young’s modulus values of various EP/PFDCHQ composites are shown in Fig. 15 and Table S2. With the increasing PFDCHQ amount, the impact strength of EP/PFDCHQ composites tends to increase firstly and then decrease, and the Young’s modulus is always higher than EP-0. The results could be explained by the introduction of rigid groups (ferrocene and benzene ring) into EP matrix. Compared to EP-0, the tensile strength and elongation at break of EP/PFDCHQ composites are both decreased a lot caused by the formation of internal stress during curing of EP [68]. The internal stress could create micro-cracks and voids to reduce the mechanical properties of materials. On the other hand, the excessive PFDCHQ is not uniformly distributed in the EP matrix, which causes the PFDCHQ acts as stress concentrators and decreases the mechanical properties of EP.


The SEM micrographs of EP-5 and EP-7 composites are shown in Fig.16. The fractured surface of

EP-5 is smooth, while the cross-sectional morphology of EP-7 has many textures and is much rougher compared to EP-5, attributing to the agglomeration of excess PFDCHQ in epoxy resin matrix, thereby the agminated PFDCHQ may act as the stress concentration point to reduce the mechanical properties of EP/PFDCHQ composites.

word image 2768

Fig. 15. The impact strength, tensile strength, elongation at break and Young’s modulus of pure EP and EP/PFDCHQ composites



word image 2769

Fig. 16. SEM micrographs of the fractured surface of EP-5 (a) and EP-7 (b) composites




In this article, a novel DOPO and ferrocene-based oligomer (PFDCHQ) was first successfully synthesized and used as flame retardant in epoxy resin. The TGA results show that the PFDCHQ can contribute improved thermal and thermo-oxidative stability at high temperature region as well as char yield to epoxy resin matrix. The flame retardance and combustion behavior of EP-0 and EP-5 were comprehensively investigated. With 5 wt% addition of PFDCHQ, the LOI value of EP-5 is increased to 32.0% and can pass the V-0 rating. But it is unfortunate that the data of cone calorimeter test is not very well. The PFDCHQ can catalyze epoxy resin to form highly graphitization aromatic char in condensed phase and improve the char residue; PFDCHQ produces PO and PO2 fragments in gaseous phase for inhibiting flame development; moreover, an obvious blowing-out effect can be observed in UL-94 test for EP-3 and EP-5. This research provides a new idea for the molecular design of phosphorus and ferrocene-containing flame retardant system.


This work was financially supported by the National Natural Science Foundation of China

(51673160; 51373140), the China Scholarship Council (201508515104), the Postgraduate

Innovation Fund Project by Southwest University of Science and Technology (17ycx015) and

Longshan Academic Talent Research Support Plan of Southwest University of Science and Technology (17LZX404; 18LZX440)



  1. X. Zhang, Q. He , H. Gu, H.A. Colorado, S. Wei, Z. Guo, Flame-retardant electrical conductive nanopolymers based on bisphenol F epoxy resin reinforced with nano polyanilines, ACS Appl.

Mater. Interfaces. 5 (2013) 898-910.

  1. Yeh J-M, Huang H-Y, Chen C-L, Su W-F, Yu Y-H. Siloxane-modified epoxy resin–clay nanocomposite coatings with advanced anticorrosive properties prepared by a solution dispersion approach. Surface and Coatings Technology. 2006;200:2753-2763.
  2. F. Lapoque, K. Redford, Curing effects on viscosity and mechanical properties of a commercial epoxy resin adhesive, Int. J. Adhes. Adhes. 22 (2002) 337-346.
  3. L. Chen, S.G. Chai, K. Liu, N.Y. Ning, J. Gao, Q.F. Liu, F. Chen, Q. Fu, Enhanced epoxy/silica composites mechanical properties by introducing graphene oxide to the interface, ACS Appl.

Mater. Inter. 4 (2012) 4398-4404.

  1. X.M. Zhao, H.V. Babu, J. Llorca, D.Y. Wang, Impact of halogen-free flame retardant with varied phosphorus’s chemical surrounding on the properties of diglycidyl ether of bisphenol-A type epoxy resin: synthesis, fire behaviour, flame-retardant mechanism and mechanical properties, RSC Adv. 6 (2016) 59226-59236.
  2. S.V. Levchik, E.D. Weil, Thermal decomposition, combustion and flame-retardancy of epoxy resins-a review of the recent literature, Polym. Int. 53 (2004) 1901-1929.
  3. Y.Z. Wang, X.T. Chen, X.D. Tang, X.H. Du, A new approach for the simultaneous improvement of fire retardancy, tensile strength and melt dripping of poly(ethylene terephthalate), J. Mate.

Chem. 13 (2003) 1248-1249.

  1. Y. Zhang, L. Chen, J.J. Zhao, H.B. Chen, M.X. He, Y.P. Ni, J.Q. Zhai, X.L. Wang, Y.Z. Wang, A phosphorus-containing PET ionomer: from ionic aggregates to flame retardance and restricted melt-dripping, Polym. Chem. 5(2014) 1982-1991.
  2. Z.B. Shao, C. Deng, Y. Tan, M.J. Chen, L. Chen, Y.Z. Wang, An efficient mono-component polymeric intumescent flame retardant for polypropylene: preparation and application, ACS Appl.

Mater. Interfaces. 6 (2014) 7363-7370.

  1. C. Deng, H. Yin, R.M. Li, S.C. Huang, B. Schartel, Y.Z. Wang, Modes of action of a monocomponent intumescent flame retardant MAPP in polyethylene-octene elastomer, Polym. Degrad. Stab. 138 (2017) 142-150.
  2. S.F. Liao, C. Deng, S.C. Huang, J.Y. Cao, Y.Z. Wang, An efficient halogen-free flame retardant for polyethylene: piperazinemodified ammonium polyphosphates with different structures, Chin. J. Polym. Sci. 34 (2016) 1339-1353.
  3. Z.M. Bai, L. Song, Y. Hu, R.K.K. Yuen, Preparation, flame retardancy, and thermal degradation of unsaturated polyester resin modified with a novel phosphorus containing acrylate, Ind. Eng. Chem. Res. 52 (2013) 12855-12864.
  4. B. Schartel, B. Perret, B. Dittrich, M. Ciesielski, J. Krämer, P. Müller, V. Altstädt, L. Zang, M. Döring, Flame retardancy of polymers: the role of specific reactions in the condensed phase, Macromol. Mater. Eng. 301 (2016) 9-35.
  5. W. Kim, D. Hoang, H. Vothi, C. Nguyen, T. Giang, H. An, J. KIm, Synthesis, flame retardancy, and thermal degradation behaviors of novel organo-phosphorus compounds derived from 9,10dihydro-9-oxa-10-phosphaphenanthrene-10-oxide (DOPO), Macromol. Res. 24 (2016) 66-73. [15] A. König, E. Kroke. Flame retardancy working mechanism of methyl-DOPO and MPPP in flexible polyurethane foam, Fire Mater. 36 (2012) 1-15.
  6. M.M. Velencoso, A. Batting, J.C. Markwart, B. Schartel, F.R. Wurm, Molecular firefightinghow modern phosphorus chemistry can help solve the flame retardancy task, Angew. Chem. Int.

Ed. (2018). DOI: 10.1002/anie.201711735.

  1. S. Yang, J. Wang, S. Huo, L.F. Cheng, M. Wang, The synergistic effect of maleimide and phosphaphenanthrene groups on a reactive flame-retarded epoxy resin system, Polym. Degrad. Stab. 115 (2015) 63-69.
  2. P.J. Chao, Y.J. Li, X.Y. Gu, D.D. Han, X.Q. Jia, M.Q. Wang, T.G. Zhou, T. Wang, Novel phosphorus–nitrogen–silicon flame retardants and their application in cycloaliphatic epoxy systems, Polym. Chem. 6 (2015) 2977-2985.
  3. B. Perret, B. Schartel, K. Stöß, M. Ciesielski, J. Diederichs, M. Döring, J. Krämer, V. Altsädt, Novel DOPO-based flame retardants in high-performance carbon fibre epoxy composites for aviation, Eur. Polym. J. 47 (2011) 1081-1089.
  4. U. Braun, U. Knoll, B. Schartel, T. Hoffmann, D. Pospiech, J. Artner, M. Ciesielski, M. Döring, R. Perez-Graterol, J.K.W. Ssndler, V. Altstädt, Novel phosphorus-containing poly(ether sulfone)s and their blends with an epoxy resin: thermal decomposition and fire retardancy, Macromol. Chem.

Phys. 207 (2006) 1501-1514.

  1. Wang, Y. Hu, L. Song, H.Y. Yang, W.Y. Xing, H.D. Lu, Synthesis and characterization of a DOPOsubstitued organophosphorus oligomer and its application in flame retardant epoxy resins, Prog. Org. Coat. 71 (2011) 72-82.
  2. X. Wang, Y. Hu, L. Song, W.Y. Xing, H.D. Lu, Preparation, mechanical properties, and thermal degradation of flame retarded epoxy resins with an organophosphorus oligomer, Polym. Bull. 67 (2011) 859-873.
  3. I.D. Carja, D. Serbezeanu, T. Vlad-Bubulac, C. Hamciuc, A. Coroaba, G. Lisa, C .G. López, M.F. Soriano, V.F. Pérez, M.D. Romero Sánchez, A straightforward, eco-friendly and cost-effective approach towards flame retardant epoxy resins, J. Mater. Chem. A. 2 (2014) 16230-16241. [24] X. Wang, Y. Hu, L. Song, W.Y. Xing, H.D. Lu, P. Lv, G.X. Jie, Flame retardancy and thermal degradation mechanism of epoxy resin composites based on a DOPO substituted organophosphorus oligomer, Polymer. 51 (2010) 2435-2445.
  4. N.N. Tian, J. Gong, X. Wen, K. Yao, T. Tang, Synthesis and characterization of a novel organophosphorus oligomer and its application in improving flame retardancy of epoxy resin, RSC Adv. 4 (2014) 17607-17614.
  5. X. Wang, S. Zhou, W. Xing, B. Yu, X. Feng, L. Song, Y. Hu, Self-assembly of Ni–Fe layered double hydroxide/graphene hybrids for reducing fire hazard in epoxy composites, J. Mater. Chem. A. 1 (2013) 4383-4390.
  6. Y. Zhang, X.A. Li, Z.P. Fang, T.R. Hull, A. Kelarakis, A.A. Stec, Mechanism of enhancement of intumescent fire retardancy by metal acetates in polypropylene, Polym. Degrad. Stab. 136 (2017) 139-145.
  7. T. Tang, X.C. Chen, X.Y. Meng, H. Chen, Y.P. Ding, Synthesis of multiwalled carbon nanotubes by catalytic combustion of polypropylene, Angew. Chem. Int. Ed. 44 (2005) 1517-1520. [29] N. Najfi-Mohajeri, G.L. Nelson, R. Benrashid, Synthesis and properties of new ferrocenemodified urethane block copolymers, J. Appl. Polym. Sci. 76 (2000) 1847-1856.
  8. W.A. Amer, L. Wang, H.J. Yu, A.M. Amin, Y. Wang, Synthesis and properties of a ferrocenebased metallomesogenic polymer containing bis(4-hydroxyoctoxyphenyl)sulfone, J. Inorg.

Organomet. Polym Mater. 22 (2012) 1229-1239.

  1. S. Senthil, P. Kannan, Ferrocene-based organophosphorus liquid–crystalline polymers: synthesis and characterization, J. Polym. Sci., Part A: Polym. Chem. 39 (2001) 2396-2403. [32] M.A. Hussein, A.M. Asiri, Organometallic ferrocene and phosphorus containing polymers synthesis and characterization, Des. Monomers. Polym. 15 (2012) 207-251. [33] K. Kishore, P. Kannan, K. Iyanar, Synthesis, characterization, and fire retardancy of ferrocene containing polyphosphate esters, J. Polym. Sci., Part A: Polym. Chem. 29 (1991) 1039-1044.

[34] S. Mehdipour-Ataei, S. Babanzadeh, New types of heat-resistant, flame-retardant ferrocenebased polyamides with improved solubility, React. Funct. Polym. 67 (2007) 883-892. [35] D.J. Liao, Q.K. Xu, R.W. McCabe, H.V. Babu, X.P. Hu, N. Pan, D.Y. Wang, T.R. Hull, Ferrocenebased nonphosphorus copolymer: synthesis, high-charring mechanism, and its application in fire retardant epoxy resin, Ind. Eng. Chem. Res. 56 (2017) 12630-41263.

  1. J.X. Tao, W.J. Xiao, Structural chemistry of organotin ferrocenecarboxylic esters I. Synthesis and spectroscopic studies on dialkyltin esters of ferrocenecarboxylic acid FcCOOH and 1,1’ferrocenedicarboxylic acid Fc(COOH)2, J. Organomet. Chem. 526 (1996) 21-24.
  2. F.W. Knobloch, W.H. Rauscher, Condensation polymers of ferrocene derivatives, J. Polym. Sci., Part A: Polym. Chem. 54 (1961) 651-656.
  3. X.D. Qian, L. Song, S.H. Jiang, G. Tang, W.Y. Xing, B.B. Wang, Y. Hu, Novel flame retardants containing 9,10-dihydro-9-oxa-10-phosphaphenanthrene-10-oxide and unsaturated bonds: synthesis, characterization, and application in the flame retardancy of epoxy acrylates, Ind. Eng. Chem. Res. 52 (2013) 7307-7315.
  4. M.S. Khan, A. Nigar, M.A. Bashir, Z. Akhter, A new ferrocene – containing polyamide prepared from an improved synthesis of 1,1’-ferrocene dicarbonyl chloride and ferrocene-based diamine, Synth. Commun. 37 (2007) 473-482.
  5. Y.X. Qi, W.H. Wu, L.J. Han, H.Q. Qu, X. Han, A.Q. Wang, J.Z. Xu, Using TG-FTIR and XPS to understand thermal degradation and flame-retardant mechanism of flexible poly(vinyl chloride) filled with metallic ferrites, J. Therm. Anal. Calorim. 123 (2015) 1263-1271.
  6. H.M. Zhu, J.H. Yan, X.G. Jiang, Y.E. Lai, K.F. Cen, Study on pyrolysis of typical medical waste materials by using TG-FTIR analysis, J. Hazard. Mater. 153 (2008) 670-676.
  7. R.K. Jian, P. Wang, W.S. Duan, J.S. Wang, X.L. Zheng, J.B. Weng, Synthesis of a novel P/N/Scontaining flame retardant and its application in epoxy resin: thermal property, flame fetardance, and pyrolysis behavior, Ind. Eng. Chem. Res. 55 (2016) 11520-11527.
  8. W.J. Liang, B. Zhao, C.Y. Zhang, R.K. Jian, D.Y. Liu, Y.Q. Liu, Enhanced flame retardancy of DGEBA epoxy resin with a novel bisphenol-A bridged cyclotriphosphazene, Polym. Degrad. Stab. 144 (2017) 292-303.
  9. Z.M. Bai, L. Song, Y. Hu, X. Gong, R.K.K. Yuen, Investigation on flame retardancy, combustion and pyrolysis behavior of flame retarded unsaturated polyester resin with a star-shaped phosphorus-containing compound, J. Anal. Appl. Pyrolysis. 105 (2014) 317-326.
  10. Y. Qiu, L.J. Qian, W. Xi, Flame-retardant effect of a novel phosphaphenanthrene/triazinetrione bi-group compound on an epoxy thermoset and its pyrolysis behaviour, RSC Adv. 6 (2016) 56018-56027.
  11. W.J. Huang, W.T. He, L.J. Long, W. Yan, M. He, S.H. Qin, J. Hu, Highly efficient flame-retardant glass-fiber-reinforced polyamide 6T system based on a novel DOPO-based derivative: Flame retardancy, thermal decomposition, and pyrolysis behavior, Polym. Degrad. Stab. 148 (2018) 26-


  1. C. Ma, B. Yu, N.N. Hong, Y. Pan, W.Z. Hu, Y. Hu, Facile synthesis of a highly efficient, halogenfree, and intumescent flame retardant for epoxy resins: thermal properties, combustion behaviors, and flame-retardant mechanisms, Ind. Eng. Chem. Res. 55 (2016) 10868-10879. [48] J.L. Wang, C. Ma, P.L. Wang, S.L. W. Cai, Y. Hu, Ultra-low phosphorus loading to achieve the superior flame retardancy of epoxy resin, Polym. Degrad. Stab. 149 (2018) 119-128.
  2. H. Ebadi-Dehaghani, S. Mehdipour-Ataei, Novel ferrocene-based organometallic poly(ether sulfone amide imide)s: preparation, characterization, and properties, J. Inorg. Organomet. Polym.

22(2011) 223-234.

  1. C.L. Dong, A. Wirasaputra, Q.Q. Luo, S.M. Liu, Y.C. Yuan, J.Q. Zhao, Y. Fu, Intrinsic flameretardant and thermally stable epoxy endowed by a highly efficient, multifunctional curing agent, Materials. 9(2016) 1008-1023.
  2. P. Wang, Z.S. Cai, Highly efficient flame-retardant epoxy resin with a novel DOPO-based triazole compound: thermal stability, flame retardancy and mechanism, Polym. Degrad. Stab. 137(2017) 138-150.
  3. W.C. Zhang, X.D. He, T.L. Song, Q.J. Jiao, R.J. Yang, Comparison of intumescence mechanism and blowing-out effect in flame-retarded epoxy resins, Polym. Degrad. Stab. 112(2015) 43-51. [53] B. Schartel, T.R. Hull, Development of fire-retarded materials—Interpretation of cone calorimeter data, Fire Mater. 31 (2007) 327-354.
  4. X.D. Qian, L. Song, B. Yu, B.B. Wang, B.H. Yuan, Y.Q. Shi, Y. Hu. K.K.Y. Richard, Novel organic–inorganic flame retardants containing exfoliated graphene: preparation and their performance on the flame retardancy of epoxy resins, J. Mater. Chem. A. 1 (2013) 6822-6830.
  5. S. Brehme, B. Schartel, J. Goebbels, O. Fischer, D. Pospiech, Y. Bykov, M. Döring, Phosphorus polyester versus aluminium phosphinate in poly(butylene terephthalate) (PBT): Flame retardancy performance and mechanisms, Polym. Degrad. Stab. 96 (2011) 875-884.
  6. N.H. Huang, J.Q. Wang, A TGA-FTIR study on the effect of CaCO3 on the thermal degradation of EBA copolymer, J. Anal. Appl. Pyrolysis. 84 (2009) 124-130.
  7. Y.Y. Dong, Z. Gui, Y. Hu, Y. Wu, S.H. Jiang, The influence of titanate nanotube on the improved thermal properties and the smoke suppression in poly(methyl methacrylate), J. Hazard. Mater. 209 (2012) 34-39.
  8. T. Mariappan, C.A. Wilkie, Flame retardant epoxy resin for electrical and electronic applications, Fire Mater. 38 (2014) 588-598.
  9. D. Wang, L. Song, K.Q. Zhou, X.J. Yu, Y. Hu, J. Wang, Anomalous nano-barrier effects of ultrathin molybdenum disulfide nanosheets for improving the flame retardance of polymer nanocomposites, J. Mater. Chem. A. 3 (2015) 14307-14317.
  10. X. Wang, S. Zhou, W.W Guo, P.L. Wang, W.Y. Xing, L. Song, Y. Hu, Renewable

Cardanol-Based Phosphate as a Flame Retardant Toughening Agent for Epoxy Resins.

ACS Sustain. Chem. Eng. 5 (2017) 3409-3416.

  1. A.H. Yang, C. Deng, H. Chen, Y.X. Wei, Y.Z. Wang. A novel Schiff-base polyphosphate ester: Highly-efficient flame retardant for polyurethane elastomer, Polym. Degrad. Stab. 144 (2017) 70-


  1. Y. Zhang, B. Yu, B.B. Wang, K.M. Liew, L. Song, C.M. Wang, Y. Hu, Highly effective P–P synergy of a novel DOPO-based flame retardant for epoxy resin, Ind. Eng. Chem. Res. 56 (2017) 12451255.
  2. S. Zhou, L. Song, Z.Z. Wang, Y. Hu, W.Y. Xing, Flame retardation and char formation mechanism of intumescent flame retarded polypropylene composites containing melamine phosphate and pentaerythritol phosphate, Polym. Degrad. Stab. 93 (2008) 1799-1806. [64] L. Zhang, Y.H. Wang, Q. Liu, X.F. Cai, Synergistic effects between silicon-containing flame retardant and DOPO on flame retardancy of epoxy resins, J. Therm. Anal. Calorim. 123 (2015) 1343-1350.

[65] M. Bugajny, S. Bourbigot, M.L. Bras, R. Delobel, The origin and nature of flame retardance in ethylene-vinyl acetate copolymers containing hostaflam AP 750, Polym. Int. 48 (1999) 264-270. [66] Z.Z. Wang, P. Lv, Y. Hu, K.L. Hu, Thermal degradation study of intumescent flame retardants by TG and FTIR: Melamine phosphate and its mixture with pentaerythritol, J. Anal. Appl.

Pyrolysis. 86 (2009) 207-214.

  1. G.H. Hsiue, S.J. Shiao, H.F. Wei, W.J. Kuo, Y.A. Sha, Novel phosphorus-containing dicyclopentadiene-modified phenolic resins for flame-retardancy applications, J. Appl. Polym. Sci. 79 (2001) 342-349.
  2. S. Gaan, S.Y. Liang, H. Mispreuve, H. Perler, R. Naescher, M. Neisius, Flame retardant flexible polyurethane foams from novel DOPO-phosphonamidate additives, Polym. Degrad. Stab. 113 (2015) 180-188.


Composites Part B 179 (2019) 107487

word image 717

Intrinsically flame retardant bio-based epoxy thermosets: A review

Xin Wang, Wenwen Guo, Lei Song


, Yuan Hu


word image 2770

Contents lists available at



Composites Part B

journal homepage:


word image 2771

State Key Laboratory of Fire Science, University of Science and Technology of China, 96 Jinzhai Road, Hefei, Anhui, 230026, PR China



Bio-based epoxy

Intrinsically flame retardant

Mechanical properties Thermal properties

At present most of epoxy thermosets is dependent on petroleum-based resources especially diglycidyl ether bisphenol A (DGEBA)-type epoxy monomers produced from epichlorohydrin (ECH) and bisphenol A (BPA). Owing to the limitation of the greenhouse gas emission, development of the bio-based epoxy thermosets is gaining increasing attention to substitute the petroleum-based ones. However, the bio-based epoxy thermosets possess similar high flammability to their petroleum-based counterparts. It is thereby necessary to endow them with flame retardancy. This review article aims to summarize the most relevant and up-to-date advances in intrinsically flame retardant bio-based epoxy thermosets. First, the approaches to synthesis of bio-based intrinsically flame retardant epoxy monomers are introduced briefly. Subsequently, the review focuses in particular on partly bio-based intrinsically flame retardant epoxy thermosets from either bio-based epoxy monomers or bio-based curing agents in terms of their flame retardant property as well as mechanical property and thermal stability. Additionally, the fully bio-based intrinsically flame retardant epoxy thermosets are also reviewed. Finally, we will provide a brief comment on opportunities and challenges for future growth of bio- based intrinsically flame retardant epoxy thermosets.


Epoxy thermosets, as one of the most important thermosets, possess many favorable performances including excellent mechanical properties, good chemical and electrical resistance, superior moisture resistivity, and high dimensional stability [1–5]. These favorable performances make epoxy thermosets suitable for various end-use applications such as Electrical & Electronics, paints and coatings, adhesives, fiber-reinforced advanced composites, etc [6–9]. According to the Business Communications Company (BCC) Research report, the global market for epoxy thermosets was valued at USD 7.0 billion in 2015 and is forecasted to reach USD 10.2 billion in 2021 at a compound annual growth rate of 6.3% between 2016 and 2021 [10]. Currently, 90% global demand of epoxy thermosets originate from petroleum-based diglycidyl ether bisphenol A (DGEBA)-type epoxy monomers, which are prepared from reaction between epichlorohydrin (ECH) and bisphenol A (BPA) in the presence of sodium hydroxide [11]. Despite of these favorable performances of DGEBA aforementioned, the production of DGEBA-type epoxy monomers is mainly dependent on fossil resources, which is unfavorable to limit greenhouse gas emission [12,13].

In this context, a large amount of efforts has been made to develop bio-based epoxy thermosets from renewable resources such as plant oil, furan, lignin, rosin, vanillin, itaconic acid, etc, to substitute petroleum- based epoxy thermosets [14,15]. These bio-based epoxy thermosets display comparable properties to their petroleum-based counterparts but eliminate the drawbacks of petroleum-based counterparts. However, like petroleum-based epoxy, bio-based epoxy thermosets suffer from high flammability, which hinder their applications in the fields of cars, trains and airplanes, construction, and electronic appliance. It is thereby imperative to impart flame retardant property to bio-based epoxy thermosets. Generally, flame retardant technology can be divided to two primary categories: additive-type and reactive-type. The former one involves the physical blending of flame retardant additives to polymers, which usually requires high loading of flame retardant additives resulting in migration and leaching of additives that could have harmful effect due to human exposure, as well as deterioration in mechanical strength and thermal stability [16]. The latter one, also known as intrinsically flame retardant technology, involves the incorporation of flame retardant monomers into polymer networks through covalent bonds, which overcomes the drawbacks of additive-type flame retardants. In this review, we are focusing on intrinsically flame retardant bio-based epoxy thermosets.

* Corresponding author.

** Corresponding author.

E-mail addresses: (L. Song), (Y. Hu).

Received 15 June 2019; Received in revised form 23 September 2019; Accepted 27 September 2019 Available online 3 October 2019

1359-8368/© 2019 Elsevier Ltd. All rights reserved.

word image 2772

Scheme 1. Synthesis of epoxy resins from phenolic compounds via glycidylation reaction.

word image 2773

Scheme 2. Synthetic pathway to obtain epoxy resins from double bond oxidation.

The objective of this review is hence to provide an overview of the field by highlighting and discussing the state-of-the art, as well as future perspectives of intrinsically flame retardant epoxy thermosets derived from renewable resources. The synthetic approaches are mentioned to optimize the design of intrinsically flame retardant bio-based epoxy thermosets. Special emphasis will be focused on partly bio-based intrinsically flame retardant epoxy thermosets from either bio-based epoxy monomers or bio-based curing agents in terms of their flame retardant property as well as mechanical property and thermal stability. Furthermore, the fully bio-based intrinsically flame retardant epoxy thermosets are also reviewed. Finally, we will provide a brief comment on opportunities and challenges for future growth of bio-based intrinsically flame retardant epoxy thermosets.

Approaches to synthesis of bio-based intrinsically flame retardant epoxy monomers

The synthetic routes of bio-based intrinsically flame retardant epoxy monomers are very similar to the traditional ones, which are summarized as follows.

From ECH

The production of epoxy monomers from ECH results from the condensation reaction between phenols and ECH using sodium hydroxide as catalyst, as diagrammatically shown in Scheme 1. According to the previous report [17], the condensation reaction between the phenate ion (ArO ) and ECH has two competitive mechanisms: (i) nucleophilic substitution (mechanism SN2) with removal of the sodium chloride; (ii) ring opening of ECH with phenate ion followed by intramolecular cyclization (SNi) of the corresponding alcoholate. In this approach, it is crucial to synthesize phenols with flame retardant elements such as phosphorus, silicon and nitrogen, followed by reaction with ECH. This approach is the most popular to design bio-based intrinsically flame retardant epoxy monomers.

From double bond oxidation

Another approach to the production of bio-based intrinsically flame retardant epoxy monomers is the peroxidation of a carbon carbon double bond (Scheme 2). The reactivity of the carbon carbon double bond differs from the chemical surroundings of the double bonds. The oxidation of the double bonds in the aliphatic chains just requires the use of hydrogen peroxide [18]. However, stronger oxidative reagents such as m-chloroperbenzoic acid (m-CPBA) are needed when the double bonds belong to glycidyl type [19]. Additionally, selective epoxidation method for internal and terminal carbon carbon double bonds was also reported by Colladon et al. [20]. Hydrogen peroxide together with a Pt (II) complex showed high selectivity in epoxidation for terminal carbon carbon double bonds, while m-CPBA only favors epoxidation for internal double bond. The production of the intrinsically flame retardant epoxy monomers derived from bio-mass with carbon carbon double bonds are usually adopted by this method.

Partly bio-based intrinsically flame retardant epoxy thermosets

Epoxy thermosets are prepared through formation of three- dimensional cross-linking networks from chemical reaction between epoxy monomers and curing agents. In this review, if the resultant epoxy thermoset is obtained from bio-based epoxy monomer plus petroleum- based curing agent or petroleum-based epoxy monomer plus bio-based curing agent, we call it partly bio-based epoxy thermoset. This section will review partly bio-based intrinsically flame retardant epoxy thermosets from either bio-based flame retardant epoxy monomers or bio- based flame retardant curing agents.

Bio-based flame retardant epoxy monomers for thermosets

This section will highlight and discuss the state-of-the-art of thermosetting epoxy monomers derived from a wide variety of renewable resources including fatty acid, eugenol, vanillin, daidzein, furan, itaconic acid, etc. According to the chemical structure, the bio-based flame retardant epoxy monomers are divided into three categories: Bio-based aromatic epoxy monomers, bio-based epoxy monomers derived from terpenes and bio-based aliphatic epoxy monomers.

Bio-based aromatic epoxy monomers

Daidzein-based epoxy monomers. Daidzein is a natural bisphenol compound that can be isolated from soybeans [21]. It has been regarded

word image 2774

Scheme 3. Synthetic route to daidzein-based epoxy monomer (DGED).

as a green product for wide applications in the food and pharmacy industry [22]. Very recently, Dai et al. synthesized a diglycidyl ether of daidzein (DGED) through glycidylation reaction between daidzein and ECH (Scheme 3) [23]. The DGED was subsequently cured by 4, 40-diaminodiphenylmethane (DDM) to obtain a thermosetting epoxy, and DGEBA cured by DDM was prepared as a comparative sample. Without the incorporation of additional flame retardants, the DGED/DDM thermosets displayed a relatively high limiting oxygen index (LOI) of 31.6% and V-0 classification in UL-94 vertical burning test, whereas the DGEBA/DDM showed a LOI of 24.5% and no classification in UL-94 vertical burning test. The superior flame retardancy of the DGED/DDM thermosets could be attributed to the excellent char-forming ability of the DGED, as evidenced by the char yield of 43% at 800 C under nitrogen. The notably enhanced Tg and mechanical properties were ascribed to the high cross-linking density of the cured DGED resulted from the dimerization of benzopyrone ring. As far as the outstanding integrated properties are concerned, daidzein is considered to be an ideal candidate for synthesis of high-performance epoxy thermosets, but the only negative outcome is the lower initial thermal degradation temperature of the DGED/DDM thermosets than that of the DGEBA/DDM thermosets.

Eugenol-based epoxy monomers. Eugenol is a major component of clove oil, which is less toxic and relatively cost-effective [24]. The chemical structure of eugenol simultaneously contains reactive phenolic hydroxyl group and allyl group, which could be used to synthesize bio-based epoxy through either glycidylation reaction with ECH or oxidation of double bond. Due to its unique structure, eugenol has gained considerable attention in the field of preparation of bio-based epoxy thermosets [15,25–30]. Among these studies, there are several eugenol-based epoxy thermosets with intrinsical flame retardancy [29–31]. A eugenol-based difunctional epoxy monomer named TPEU-EP was synthesized by Wan et al. through two steps (Scheme 4a), and then cured by 3, 30-diaminodiphenyl sulfone (33DDS) [29]. The TPEU-EP/33DDS system revealed a higher LOI of 26.8% than the DGEBA/33DDS (23.5%). In the microscale combustion calorimeter (MCC) measurements, the peak heat release rate (PHRR) and the total heat release (THR) of the TPEU-EP/33DDS were decreased by 68% and 40%, respectively, compared to those of the DGEBA/33DDS. In the UL-94 vertical burning test, the TPEU-EP/33DDS could self-extinguish in 24 s, while the DGEBA/33DDS burned out finally (Fig. 1), indicating intrinsic flame retardancy of the TPEU-EP/33DDS. In Wan’s next work [30], another eugenol-based difunctional epoxy monomer named DEU-EP was synthesized (Scheme 4a) and cured by DDM. The PHRR and THR of the DEU-EP/DDM were reduced by 55% and 38%, respectively, as compared to those of the DGEBA/DDM in the MCC measurements. As well, the DEU-EP/DDM could self-extinguish in 10 s, whereas the DGEBA/DDM could not in the horizontal burning test. However, in Wan’s studies, there is still space to further improve the performances of

word image 2775

Scheme 4. Schematic for the synthesis route of various eugenol-based epoxy monomers.

word image 2776Fig. 1. Snapshots of the vertical burning test of DGEBA/33DDS and TPEU-EP/33DDS. The samples were ignited at the bottom by a Bunsen burner for 10 senconds.

The burnt residues are compared in the bottom right image [24].

eugenol-based epoxy thermosets, like failure in UL-94 V-0 classification as well as deterioration in thermal stability in terms of lower Tg and initial thermal degradation temperature.

In order to further improve the flame retardancy of the eugenol- based epoxy thermosets, flame retardant elements like phosphorus and silicon are considered to incorporate into eugenol-based epoxy monomers. Faye and co-workers synthesized an eugenol-based epoxy monomer, tri(epoxized-eugenyl)phosphate (TEEP) (Scheme 4b) [31].

word image 2777

Scheme 5. (a) Synthesis of a difunctional furan-based epoxy monomer named EUFU-EP; (b) Chemical structures of furan-based epoxy monomers: BOF and OmbFdE, and different amines: DEGA, TEGA, and TGDE.

The TEEP cured by meta-xylylenediamine (MXDA) and 2,2’-(Ethane-1, 2-diylbis(oxy)) bis(ethan-1-amine) (EDR-148) displayed a Tg of 84 C and 62 C, respectively, whereas the DGEBA/MXDA and the DGEBA/EDR-148 exhibited a much higher Tg of 116 C and 98 C. The presence of the methoxy groups in the TEEP was believed to induce undesirable effect on the Tg. Due to the presence of the phosphate esters structure, the TEEP/MXDA and the TEEP/EDR-148 also showed lower thermal stability than their DGEBA counterparts. But a much higher char

word image 2778

Fig. 2. Heat release rate versus time plots from MCC tests for the cured EUFU- EP/MHHPA and DGEBA/MHHPA thermosets [33].

yield at 600 C was observed for the TEEP/MXDA (33%) and the TEEP/EDR-148 (36%) in TGA test. Such a high char yield is favorable for improving flame retardancy. In another work, Miao et al. also synthesized the TEEP thermosets cured by DDM [32]. The LOI increased to 31.4% and the UL-94 vertical burning behavior enhanced to V-0 classification for the TEEP/DDM system. In the MCC measurements, the PHRR and the THR of the TEEP/DDM were 63.1% and 57.4% lower than those of the DGEBA/DDM. Furthermore, the TEEP/DDM presented lower dielectric constant and dielectric loss than the DGEBA/DDM, which is suitable for electrical and electronic applications. Very recently, Li et al. reported three kinds of silicone-containing eugenol-based epoxy monomers (SIEEP2, SIEEP4, and SIEPEP) (Scheme 4c) [33]. The LOI of the DGEBA cured by 4, 40-diamino diphenyl sulfone (DDS) was 22.8%, whereas that for the SIEEP2/DDS, SIEEP4/DDS and SIEPEP/DDS was 26.0%, 28.0% and 31.0%, respectively. Although the

PHRR and the THR of the SIEEP2/DDS, the SIEEP4/DDS, and the SIEPEP/DDS measured by cone calorimeter decreased obviously compared to those of the DGEBA/DDS, the time to ignition (TTI) shifted to lower temperature which was undesirable for flame retardant materials. These silicone-containing eugenol-based epoxy monomers also displayed a relatively lower viscosity (<2.5 Pa s) than the DGEBA (10.7 Pa s) which was attractive for composites and prepregs applications. Ecochard et al. synthesized three kinds of phosphorylated eugenol-based epoxy thermoset (TEEP, DEEP and DEEP-Ph, Scheme 4b and 4d) cured by MXDA [34]. In the MCC measurements, although the PHRR and the THR of the TEEP, DEEP and DEEP-Ph thermosets were reduced significantly compare to those of the DGEBA, no significant difference was observed among TEEP, DEEP and DEEP-Ph thermosets, suggesting phosphate and phosphonate groups contributed equally to promote char yield and reduce flammability.

Furan-based epoxy monomers. Owing to a lot of merits including high bio-safety, ready availability and desirable biodegradability, furan- based chemicals have been identified as the top value-added building blocks for technology development by the U.S. Department of Energy [35]. Furan-based chemicals have also been considered to potentially substitute petroleum-based phenyl building blocks in synthesis of thermosetting materials because of their aromatic structures. Over the past few decades, furan-based chemicals have widely been used as starting materials for synthesis of bio-based epoxy thermosets [36,37]. However, only a limited number of furan-based epoxy thermosets focus on flame retardancy. Miao et al. synthesized a bis

(2-methoxy-4-(oxiran-2-ylmethyl)phenyl) furan-2,5-dicarboxylate (EUFU-EP) (Scheme 5a) [38]. The MCC tests manifested that both the PHRR and the THR of the EUFU-EP cured by methyl hexahydrophthalic anhydride (MHHPA) were 19.0% lower than those of the DGEBA/MHHPA (Fig. 2), which was attributed to the higher char yield. The highly compact aromatic rings in EUFU-EP accounted for the higher char yield that induced condensed-phase flame retardant mechanism. The dynamic mechanical analysis (DMA) measurements indicated that the cured EUFU-EP/MHHPA thermoset had higher Tg (153 C) than the DGEBA/MHHPA thermoset (144 C), owing to the presence of the rigid rod-like aromatic ester and furan structures in EUFU-EP. Similar to other bio-based epoxy thermosets, however, the cured EUFU-EP/MHHPA thermoset exhibited lower initial degradation temperature than the DGEBA/MHHPA thermoset, which is unfavorable for applications that require high thermal resistance.

word image 2779

Fig. 3. Snapshots recorded for the burning test of the BOF/DEGA, the BOF/TEGA and the OmbFdE/TEGA [34].

Another two furan-based epoxy monomers, 2,5-bis[(2-oxiranylmethoxy) methyl]furan (BOF) and bis-furan diepoxide (OmbFdE), were reported very recently, as illustrated in Scheme 5b [39]. The BOF and the OmbFdE were cured with three different curing agents: 2, 20-oxybis (ethan-1-amine) (DEGA), 2, 2’-(ethane-1,2-diylbis(oxy))bis(ethan-1-amine) (TEGA), and 2, 2’-((oxybis(ethane-2,1-diyl))bis(oxy))bis

word image 2780

Scheme 6. Diagrammatical illustration showing the synthesis of two vanillin- based epoxy monomers called EP1 and EP2.

(ethan-1-amine) (TGDE) (Scheme 5b). Comparison between the OmbFdE/amine and the BOF/amine was studied to evaluate the effect of both furan cores and the ether motifs on the thermal property and the flame retardancy of the thermosets. In the burning test, the BOF/DEGA and the BOF/TEGA burnt out, while the OmbFdE/TEGA can self-extinguish within 30 s (Fig. 3). DSC thermograms indicated that the OmbFdE/amine and the BOF/amine showed a relatively low Tg (7–25 C), which may allow their potential application as thermo-responsive polymers between ambient and body temperatures.

Vanillin-based epoxy monomers. Vanillin is a non-toxic and renewable product which can be produced from depolymerization of lignin. It contains an aromatic ring structure with hydroxyl and aldehyde functional groups, which enables it to be served as a fascinating building block for preparation of bio-based epoxy thermosets [40–42]. Until now, vanillin-based epoxy with intrinsic flame retardancy has been rarely reported. Wang et al. synthesized two vanillin-based epoxy monomers (EP1 and EP2), as illustrated in Scheme 6 [42]. The cured EP1/DDM and EP2/DDM showed excellent flame retardancy compared to the cured DGEBA/DDM. Specifically, the LOI value for the EP1/DDM and the EP2/DDM was 31.4% and 32.8%, respectively, while that for the DGEBA/DDM was 24.6%; moreover, both the EP1/DDM and the

EP2/DDM achieved UL-94 V-0 classification. The authors attributed the excellent flame retardancy to the intumescent and dense char formation ability of the EP1/DDM and the EP2/DDM (Fig. 4). Additionally, the EP1/DDM and the EP2/DDM possessed relatively high Tg value of 183 and 214 C, much higher than the cured DGEBA/DDM with Tg of 166 C. The properties of vanillin-based epoxy thermosets could be easily tailored by adjusting the chemical structures of diamines serving as coupling agent in the synthesis process.

word image 2781

Fig. 4. (a) Digital photographs and (b) SEM microimages of the char residues after LOI test [37].

Cardanol-based epoxy monomers. Cardanol is a cost-effective and available by-product of the cashew industry, which could be used as a versatile platform for bio-based polymers and additives [43]. Ecochard et al. reported a phosphorylated cardanol based epoxy thermoset (TECP, Scheme 7) cured by MXDA [34]. As observed by cone calorimeter test, the TECP/MXDA thermoset showed a PHRR of 708 kW/m2, which was much lower than the DGEBA/MXDA (1486 kW/m2), but its THR (20.4 kJ/g) was very close to the DGEBA/MXDA (22.1 kJ/g), which was assigned to the aliphatic chains of cardanol. Undoubtedly, these aliphatic chains consisted of a high amount of carbon and hydrogen, contributing to higher heat release. It was also found that the phosphorylated eugenol based epoxy exhibited superior flame retardancy over the phosphorylated cardanol based one.

Bio-based epoxy monomers derived from terpenes

Rosin-based epoxy monomers. As an abundant and natural product, rosin is produced by heating fresh tree resin to remove the volatile liquid terpenes [28]. The production of rosin is approximately 1.2 million tons every year [44]. It has rigid hydrogenated phenanthrene ring in the molecular structure which makes it suitable as an alternative

word image 2782

Scheme 7. Synthetic route to a cardanol-based epoxy monomer (TECP).

word image 2783

Scheme 8. Synthetic route to a rosin-based siloxane epoxy monomer (AESE).

Table 1

Effect of the AP-EGDE/PMPS ratio on the LOI and the residue of the cured samples [42].



LOI (%)

Residue at 700 C (%)a
























a Determined by TGA, air atmosphere, 10 C/min; b: PAESE means physical

mixture of AP-EGDE and PMPS.

to DGEBA. Although the use of rosin and its derivatives in preparation of bio-based epoxy thermosets has been extensively investigated [15,45, 46], the flame retardant rosin-based epoxy thermosets has rarely reported up to now. One example is a rosin-based siloxane epoxy monomer (AESE) that was prepared by the reaction of ethylene glycol diglycidyl ether modified acrylpimaric acid (AP-EGDE) with polymethylphenylsiloxane (PMPS) (Scheme 8) [47]. Table 1 lists the LOI value of the samples. It can be seen that the incorporation of PMPS resulted in the improvement in the LOI value compared to the AP-EGDE/MHHPA thermosets. The highest LOI value of 30.2% was observed for the AESE30/MHHPA (the number represented the PMPS weight percentage in the AESE) among the samples. Note that the char yield for both the AESE40/MHHPA and the AESE50/MHHPA was much higher than that for the AESE30/MHHPA, but their LOI value was lower than that of the AESE30/MHHPA. The authors attributed this abnormal phenomenon to that the initial decomposition temperature for the AESE40 and the AESE50 was too high to form a protective layer. Moreover, the LOI value of AESE30/MHHPA (chemical bonding between AP-EGDE and PMPS) was higher than that of PAESE30/MHHPA (physical mixture of AP-EGDE and PMPS). Owing to the presence of the flexible chains of the PMPS, all the AESE/MHHPA thermosets displayed a relatively low tensile strength (<15 MPa) but a much larger breaking elongation (>50%), demonstrating good toughening effect of the PMPS in the AESE monomer. Thereby, it is suitable to be used as a modifier for petroleum-based epoxy monomers like DGEBA that need to improve toughness and flame retardancy simultaneously.

Bio-based aliphatic epoxy monomers

Fatty acid-based epoxy monomers. Fatty acids are the major components of the triglyceride oils, which possess carbon carbon

word image 2784

Scheme 9. Synthetic route to fatty acid-based epoxy monomer (DOPO-III) and chemical structures of UDTGE, UDBME, DDM and BAMPO.

Table 2

Formulations and LOI values of the fatty acid-based epoxy thermosets [43,44].


Molar ratio of epoxy monomers

Phosphorus content (%)

LOI (%)




/ 1/1a








/ 1/1b


















double bonds in their chemical structures [48]. These double bonds could be used as reactive sites for modifications. Epoxidation is one of the most important modifications of these double bonds leading to fatty acid-based epoxy monomers. Lligadas et al. reported a novel phosphorous-containing fatty acid-based epoxy monomer (DOPO-III) and subsequently cured with DDM and bis(m-aminophenyl) methylphosphine oxide (BAMPO) (Scheme 9) [49]. The LOI values of the DOPO-III/DDM and the DOPO-III/BAMPO thermosets were 31% and 32%, respectively, implying good flame retardancy. In another study [48], Lligadas et al. synthesized two fatty acid derived epoxy monomers: epoxidized 10-undecenoyl triglyceride (UDTGE) and epoxidized methyl 3,4,5-tris(10-undecenoyloxy)benzoate (UDBME) (Scheme 9). The UDTGE or the UDBME was combined with the DOPO-III and cured by DDM or BAMPO. The LOI values of the cured thermosets are listed in Table 2. It can be seen that the addition of DOPO-III increased the LOI of the cured thermosets obviously. The improved flame retardancy is attributable to the formation of a protective phosphorous-rich char layer that inhibits the combustible volatiles from the degraded polymer to fuel the flame. Unfortunately, because of the flexibility of long chain of fatty acids, the Tg values of the DOPO-III/DDM and the DOPO-III/BAMPO thermosets were 108 and 95 C [49], respectively, which were much lower than those of the DGEBA-based thermosets as reported previously. As a result, fatty acid-based epoxy monomers remain limited to non-structural applications like paints and coatings.

Itaconic acid-based epoxy monomers. Itaconic acid (ITA) is produced from the fermentation of carbohydrate (like glucose) in the presence of Aspergillus terreus [50]. It has two carboxyl groups as well as one carbon carbon double bond in the molecular structure, which makes it greatly suitable as a versatile platform for preparation of polymeric materials. Thereby, ITA has been regarded as one of the top twelve potential bio-based platform chemicals by the US Department of Energy [51]. Ma and co-workers synthesized a phosphorus-containing ITA-based epoxy monomer (EADI) (Scheme 10) [50]. The

EADI/MHHPA showed intrinsic flame retardancy with a V-0 classification in the UL-94 vertical burning test, whereas the DGEBA/MHHPA burned out finally. Undoubtedly, the improvement in flame retardancy was attributable to the presence of phosphorus-containing structure in EADI, but the thermal stability in terms of the initial degradation temperature was deteriorated as well.

Sebacic acid-based epoxy monomers. Sebacic acid (SA), also known as 1,10-decanedioic acid, is normally made from castor oil. Owing to the presence of two carboxyl groups in the molecular structure, SA has been widely used in production of engineering plastics like polyester [52] and nylon [53,54]. A phosphorus-containing SA-derived epoxy monomer (PSAE) was recently synthesized (Scheme 11), and mixed with a petroleum-based epoxy monomer (CE) at different ratios [55]. These mixtures were subsequently cured by polyamide hardener. The LOI value increased gradually with the increase of the PSAE loading in the thermosetting system. The LOI value reached 27% for the PSAE80/CE20 formulation (the number represents weight percentage). In the UL-94 vertical burning test, the self-extinguishing phenomenon within a period of 5–10 s was observed for the formulations of PSAE40/CE60, PSAE60/CE40 and PSAE80/CE20. Moreover, no dripping phenomenon was observed, suggesting good structural stability of the residual char. However, the SA has long aliphatic chains similar to the fatty acid, which induces low Tg and strength. As a consequence, such phosphorus-containing SA-based epoxy monomers remain limited to flame retardant coating applications.

Based on these bio-based flame retardant epoxy monomers aforementioned in this section, several phosphorus-containing bio-based epoxy monomers are selected to compare their flame retardant efficiency, as summarized in Table 3. It can be seen that char yield is roughly proportional to phosphorus content if epoxy monomers and curing agents have similar structures. Generally, higher char yield results in better flame retardancy for polymers [56]. If taking the chemical structures into account, epoxy monomers and curing agents with more aromatic structures is favorable to higher char yield, while those with aliphatic structures seems to generate lower char yield. In addition, epoxy thermosets with more aromatic structures possess high rigidity and Tg which are suitable for heat resistant thermosets applications, whereas those with aliphatic structures exhibit excellent flexibility which could serve as toughening agents.

Bio-based flame retardant curing agents for epoxy thermosets

Generally, the curing agents for epoxy thermosets include amines, amides, anhydrides, phenols, and polyphenols [57]. These curing agents contain reactive functional groups such as amino, hydroxyl and carboxyl groups, which can react with epoxide group to form three-dimensional cross-linking networks. This section will review bio-based flame retardant curing agents. However, only a few studies have been reported for bio-based flame retardant curing agents.

word image 2785

Scheme 10. Schematic representation of synthesis of a phosphorus-containing ITA-based epoxy resin (EADI).

word image 2786

Scheme 11. Schematic representation of synthesis of a phosphorus-containing sebacic acid-derived epoxy monomer called PSAE.

Mao and co-workers reported a castor oil-based binary acid as flame retardant epoxy curing agent (IDDRA) (Scheme 12) [58]. The IDDRA was combined with methyl nadic anhydride (MNA) as co-curing agent for DGEBA. The LOI, tensile property and Tg of the samples are listed in Table 4. Although the LOI value increased gradually with the increase of the IDDRA loading, both the tensile strength and the Tg dropped significantly. The elongation at break of the epoxy thermosets increased obviously as the IDDRA loading increased, indicating good toughening effect which originated from the long aliphatic chains of the castor oil species in the IDDRA. The low tensile strength and the low Tg restricted the application of the IDDRA in high performance epoxy thermosets.

Guo et al. synthesized a cardanol-based phosphorus-containing benzoxazine (CBz) as a curing agent for DGEBA, as depicted in Scheme 13 [59]. The incorporation of 15 wt% CBz into DGEBA exhibited a high LOI of 32% as well as UL-94 V-0 rating, but the reduction in the PHRR and the THR was limited, and the thermal stability was lowered. Thereby, CBz was combined with boron doped graphene (BG) nanosheets to improve the flame retardant efficiency and the thermal stability of the epoxy thermosets. With the addition of 13 wt% CBz and 2 wt % BG, the resulting DGEBA thermoset exhibited 48% and 12% lower PHRR and THR than the un-modified DGEBA thermoset, respectively. Furthermore, the impact strength of the DGEBA thermoset with 13 wt%

Table 3

Comparison of flame retardant property of several phosphorus-containing bio-based epoxy monomers.

Epoxy monomers

Curing agents

P content (%)

Char yield (%)a

LOI (%)


PHRR reduction (%)b










































































a Measured by TGA under nitrogen; b: Measured by cone calorimeter under 35 kW/m2.

word image 2787

Scheme 12. Synthetic route to a castor oil-based curing agent named IDDRA.

Table 4

The formulations, LOI and tensile property of the cured samples [53].




LOI (%)

Tensile strength (MPa)

Elongation at break (%)































CBz and 2 wt% BG increased by 22% relative to that of the un-modified DGEBA thermoset. The improved flame retardant efficiency could be attributed to that the presence of BG could improve the resistance to thermal oxidative degradation of the char layers, and the enhanced toughness was assigned to the flexible aliphatic chains of cardanol moieties.

Another two phosphorus-containing vanillin-based curing agents (TP and MP) (Scheme 14) were reported for flame retardant DGEBA thermosets very recently [60]. The TP and MP were used as co-flame retardants with ammonium polyphosphate (APP) in the DGEBA/DDM thermosets. With 10 phr of APP, the DGEBA/DDM thermoset showed no classification in UL-94 burning test. By contrast, when TP or MP was combined with APP, the resulting DGEBA/DDM system passed UL-94 V-0 classification and the LOI value was around 29%. The Tg of the epoxy thermosets with the co-addition of TP or MP with APP was comparable to or even higher than the control sample, while that for APP alone-modified epoxy thermoset compromised a lot. Thereby, these vanillin-based curing agents are suitable to be used in fabrication of high-performance epoxy thermosets.

word image 2788

Scheme 13. Diagrammatical illustration of synthesis of a cardanol-based phosphorus-containing benzoxazine (CBz).

word image 2789

Scheme 14. Chemical structures of TP, MP, MAPDGR and MMDOPO.

word image 2790

Fig. 5. The tensile strain-stress curves of cured DGEBA thermosets with different MMDOPO/MAPDGR weight ratios [56].

Yang et al. reported two bio-based phosphorus-containing curing agents (MMDOPO and MAPDGR) (Scheme 14) [61]. Owing to the unique chemical structure, the DGEBA/MMDOPO was a brittle material, whereas the DGEBA/MAPDGR was very flexible, as shown in Fig. 5. Thereby, these two curing agents were mixed with different weight ratios to form the cured DGEBA thermosets with balanced properties. This work provides a solution to tailor comprehensive properties of the cured epoxy thermosets by combining different bio-based curing agents.

Fully bio-based flame retardant epoxy thermosets

Fully bio-based flame retardant epoxy thermosets are defined as bio- based epoxy monomers cured by bio-based curing agent in this review. Owing to the increasing concern on sustainable development, the biomass content in the materials is expected as high as possible. Given this, it is desirable to develop fully bio-based flame retardant epoxy thermosets. This section will discuss the fully bio-based epoxy thermosets with intrinsic flame retardancy.

Menard and co-workers prepared a fully bio-based flame retardant epoxy thermoset through phloroglucinol-based epoxy monomers (P3EP and P2EP1P) and furan-based curing agent (DIFFA) (Scheme 15) [62]. For comparison, DGEBA and petroleum-based curing agents (IPDA and

word image 2791

Scheme 15. Chemical structures of P3EP, P2EP1P, DIFFA, DA10 and IPDA.

word image 2792

Fig. 6. Heat release rate versus time plots from MCC tests for the cured P3EP/ IPDA, P3EP/DIFFA, P3EP/DA10 and DGEBA/IPDA thermosets [57].

DA10) (Scheme 15) were also used to fabricate the control samples. The MCC results demonstrated that the flammability of the epoxy thermosets was mainly dominated by the charring ability of the epoxy monomers and the curing agents (Fig. 6). Generally, the charring ability was affected by the aromaticity and the phosphorus content of materials. Thereby, the fully bio-based epoxy thermoset based on P3EP and DIFFA showed much lower PHRR (119 W/g) than partly bio-based epoxy thermoset (395 W/g for P3EP/IPDA and 185 W/g P3EP/DA10) and petroleum-based epoxy thermoset (664 W/g for DGEBA/IPDA). The co-addition of P3EP and P2EP1P further decreased the PHRR value of the epoxy thermoset. However, the incorporation of P2EP1P into the epoxy thermosets led to a decreased Tg because of the decreased cross-linking density. The introduction of P2EP1P also decreased the thermal stability of the epoxy thermoset, but increased the char yield that accounted for the enhanced flame retardancy.

Summary and perspective

With the increasing awareness about sustainable development and environmental protection, bio-based epoxy thermosets made from various renewable products including fatty acid, eugenol, vanillin, daidzein, furan, itaconic acid, rosin, lignin, tannin, vegetable oils, etc, have attracted extensive attention to substitute petroleum-based counterparts over the past decades. Like most of synthetic polymeric materials, bio-based epoxy thermosets encounter the serious drawback of high flammability. It is thereby imperative to endow them with high flame retardancy. Up to now, intrinsically flame retardant bio-based epoxy thermosets have not widely reported, suggesting that the research on this field is just beginning. Generally, intrinsically flame retardant bio-based epoxy thermosets can be designed by two approaches: (i) selecting bio-based chemicals with high charring ability and abundant aromatic structures; (ii) incorporating flame retardant elements such as phosphorus, silicon, boron, etc into the molecular structure of bio-based epoxy monomers or curing agents.

Currently, although very high flame retardancy has been achieved for several bio-based epoxy thermosets, the thermal stability, the glass transition temperature and the strength are found to be inferior to their DGEBA counterparts. Additionally, the relatively high cost of bio-based materials for synthesis of bio-based epoxy thermosets restricts the scalable production. Owing to these problems, there is still a long way for commercialization of bio-based epoxy thermosets in replacement of the DGEBA ones. Considering the development trend in this field, increasing functionality of bio-based epoxy monomers or curing agents, improving bio-mass content and bio-degradation ability are several key factors for developing high performance bio-based epoxy thermosets in the future.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.


We gratefully acknowledge financial support from the National Natural Science Foundation of China (Grant No.: 21604081).


  1. Jin FL, Li X, Park SJ. Synthesis and application of epoxy resins: a review. J Ind Eng Chem 2015;29:1–11.
  2. Mohan P. A Critical Review: the modification, properties, and applications of epoxy resins. Polym-Plast Technol 2013;52:107–25.
  3. Kumar S, Krishnan S, Mohanty S, Nayak SK. Synthesis and characterization of petroleum and biobased epoxy resins: a review. Polym Int 2018;67:815–39.
  4. Wang X, Hu Y, Song L, Xing WY, Lu HD, Lv P, Jie GX. Flame retardancy and thermal degradation mechanism of epoxy resin composites based on a DOPO substituted organophosphorus oligomer. Polymer 2010;51:2435–45.
  5. Wang X, Hu Y, Song L, Xing WY, Lu HD. Thermal degradation behaviors of epoxy resin/POSS hybrids and phosphorus-silicon synergism of flame retardancy. J Polym Sci, Polym Phys Ed 2010;48:693–705.
  6. Fu XL, Wang X, Xing WY, Zhang P, Song L, Hu Y. Two-dimensional cardanol- derived zirconium phosphate hybrid as flame retardant and smoke suppressant for epoxy resin. Polym Degrad Stab 2018;151:172–80.
  7. Qiu SL, Wang X, Yu B, Feng X, Mu X, Yuen RKK, Hu Y. Flame-retardant-wrapped polyphosphazene nanotubes: a novel strategy for enhancing the flame retardancy and smoke toxicity suppression of epoxy resins. J Hazard Mater 2017;325:327–39.
  8. Qiu SL, Ma C, Wang X, Zhou X, Feng XM, Yuen RKK, Hu Y. Melamine-containing polyphosphazene wrapped ammonium polyphosphate: a novel multifunctional organic-inorganic hybrid flame retardant. J Hazard Mater 2018;344:839–48.
  9. Zhou JJ, Xu KJ, Xie MJ, Wu HF, Hua Z, Wang ZK. Two Strategies to precisely tune the mechanical properties of plant oil-derived epoxy resins. Compos B Eng 2019; 173:106885.
  10. Kumar A. Epoxy resins: applications and global markets. Report from BCC Inc.

Pub.; 2017. PLS077B.

  1. Shibata M, Ohkita T. Fully biobased epoxy resin systems composed of a vanillin- derived epoxy resin and renewable phenolic hardeners. Eur Polym J 2017;92:


  1. Moudood A, Rahman A, Khanlou HM, Hall W, Ochsner A, Francucci G. Environmental effects on the durability and the mechanical performance of flax fiber/bio-epoxy composites. Compos B Eng 2019;171:284–93.
  2. Cicala G, Pergolizzi E, Piscopo F, Carbone D, Recca G. Hybrid composites manufactured by resin infusion with a fully recyclable bioepoxy resin. Compos B Eng 2018;132:69–76.
  3. Boccarusso L, Durante M, Langella A. Lightweight hemp/bio-epoxy grid structure manufactured by a new continuous process. Compos B Eng 2018;146:165–75.
  4. Kumar S, Samal SK, Mohanty S, Nayak SK. Recent development of biobased epoxy resins: a review. Polym-Plast Technol 2018;57:133–55.
  5. Wang X, Kalali EN, Wan JT, Wang DY. Carbon-family materials for flame retardant polymeric materials. Prog Polym Sci 2017;69:22–46.
  6. Auvergne R, Caillol S, David G, Boutevin B, Pascault JP. Biobased thermosetting epoxy: present and future. Chem Rev 2014;114:1082–115.
  7. Hidalgoa P, Navia R, Hunter R, Gonzalez ME, Echeverría A. Development of novel bio-based epoxides from microalgae Nannochloropsis gaditana lipids. Compos B Eng 2019;166:653–62.
  8. Aouf C, Nouailhas H, Fache M, Caillol S, Boutevin B, Fulcrand H. Multi- functionalization of gallic acid. Synthesis of a novel bio-based epoxy resin. Eur Polym J 2013;49:1185–95.
  9. Colladon M, Scarso A, Sgarbossa P, Michelin RA, Strukul G. Regioselectivity and diasteroselectivity in Pt(II)-mediated “green” catalytic epoxidation of terminal alkenes with hydrogen peroxide: Mechanistic insight into a peculiar substrate selectivity. J Am Chem Soc 2007;129:7680–9.
  10. Liggins J, Bluck LJC, Runswick S, Atkinson C, Coward WA, Bingham SA. Daidzein and genistein content of fruits and nuts. J Nutr Biochem 2000;11:326–31.
  11. Kris-Etherton PM, Hecker KD, Bonanome A, Coval SM, Binkoski AE, Hilpert KF, Griel AE, Etherton TD. Bioactive compounds in foods: their role in the prevention of cardiovascular disease and cancer. Am J Med 2002;113:71–88.
  12. Dai J, Peng Y, Teng N, Liu Y, Liu C, Shen X, Mahmud S, Zhu J, Liu X. High- performing and fire-resistant biobased epoxy resin from renewable sources. ACS Sustainable Chem Eng 2018;6:7589–99.
  13. Lochab B, Shukla S, Varma IK. Naturally occurring phenolic sources: monomers and polymers. RSC Adv 2014;4:21712–52.
  14. Qin J, Liu H, Zhang P, Wolcott M, Zhang J. Use of eugenol and rosin as feedstocks for biobased epoxy resins and study of curing and performance properties. Polym Int 2014;63:760–5.
  15. Chen B, Wang F, Li JY, Zhang JL, Zhang Y, Zhao HC. Synthesis of eugenol bio- based reactive epoxy diluent and study on the curing kinetics and properties of the epoxy resin system. Chin J Polym Sci 2019;37:500–8.
  16. Liu T, Hao C, Wang L, Li Y, Liu W, Xin J, Zhang J. Eugenol-derived biobased epoxy:

shape memory, repairing, and recyclability. Macromolecules 2017;50:8588–97.

  1. Ng F, Couture G, Philippe C, Boutevin B, Caillol S. Bio-based aromatic epoxy monomers for thermoset materials. Molecules 2017;22:149.
  2. Wan JT, Gan B, Li C, Molina-Aldareguia J, Li Z, Wang X, Wang DY. A novel biobased epoxy resin with high mechanical stiffness and low flammability:

synthesis, characterization and properties. J Mater Chem 2015;3:21907–21.

  1. Wan JT, Gan B, Li C, Molina-Aldareguia J, Kalali EN, Wang X, Wang DY. A sustainable, eugenol-derived epoxy resin with high biobased content, modulus, hardness and low flammability: synthesis, curing kinetics and structure-property relationship. Chem Eng J 2016;284:1080–93.
  2. Faye I, Decostanzi M, Ecochard Y, Caillol S. Eugenol bio-based epoxy thermosets: from cloves to applied materials. Green Chem 2017;19:5236–42.
  3. Miao JT, Yuan L, Guan Q, Liang G, Gu A. Biobased epoxy resin derived from eugenol with excellent integrated performance and high renewable carbon content. Polym Int 2018;67:1194–202.
  4. Li C, Fan H, Aziz T, Bittencourt C, Wu L, Wang DY, Dubois P. Biobased epoxy resin with low electrical permissivity and flame retardancy: from environmental friendly high-throughput synthesis to properties. ACS Sustainable Chem Eng 2018;6:


  1. Ecochard Y, Decostanzi M, Negrell C, Sonnier R, Caillol S. Cardanol and eugenol based flame retardant epoxy monomers for thermostable networks. Molecules 2019;24:1818.
  2. Bozell JJ, Petersen GR. Technology development for the production of biobased products from biorefinery carbohydrates—the US Department of Energy’s “Top 10” revisited. Green Chem 2010;12:539–54.
  3. Hu F, La Scala JJ, Sadler JM, Palmese GR. Synthesis and characterization of thermosetting furan-based epoxy systems. Macromolecules 2014;47:3332–42.
  4. Shen X, Liu X, Wang J, Dai J, Zhu J. Synthesis of an epoxy monomer from bio-based 2,5-furandimethanol and its toughening via Diels–Alder reaction. Ind Eng Chem Res 2017;56:8508–16.
  5. Miao JT, Yuan L, Guan Q, Liang G, Gu A. Biobased heat resistant epoxy resin with extremely high biomass content from 2,5-furandicarboxylic acid and eugenol. ACS Sustainable Chem Eng 2017;5:7003–11.
  6. Meng J, Zeng Y, Zhu G, Zhang J, Chen P, Cheng Y, Fang Z, Guo K. Sustainable bio- based furan epoxy resin with flame retardancy. Polym Chem 2019;10:2370–5.
  7. Hernandez ED, Bassett AW, Sadler JM, La Scala JJ, Stanzione JF. Synthesis and characterization of bio-based epoxy resins derived from vanillyl alcohol. ACS Sustainable Chem Eng 2016;4:4328–39.
  8. Savonnet E, Grau E, Grelier S, Defoort B, Cramail H. Divanillin-based epoxy precursors as DGEBA substitutes for biobased epoxy thermosets. ACS Sustainable Chem Eng 2018;6:11008–17.
  9. Wang S, Ma S, Xu C, Liu Y, Dai J, Wang Z, Liu X, Chen J, Shen X, Wei J, Zhu J. Vanillin-derived high-performance flame retardant epoxy resins: facile synthesis and properties. Macromolecules 2017;50:1892–901.
  10. Caillol S. Cardanol: a promising building block for biobased polymers and additives. Curr Opin Green Sustain Chem 2018;14:26–32.
  11. Ma S, Li T, Liu X, Zhu J. Research progress on bio-based thermosetting resins.

Polym Int 2016;65:164–73.

  1. Deng L, Ha C, Sun C, Zhou B, Yu J, Shen M, Mo J. Properties of bio-based epoxy resins from rosin with different flexible chains. Ind Eng Chem Res 2013;52:


  1. Atta AM, Mansour R, Abdou MI, Sayed AM. Epoxy resins from rosin acids: synthesis and characterization. Polym Adv Technol 2004;15:514–22.
  2. Deng L, Shen M, Yu J, Wu K, Ha C. Preparation, characterization, and flame retardancy of novel rosin-based siloxane epoxy resins. Ind Eng Chem Res 2012;51:


  1. Lligadas G, Ronda JC, Galia M, Cadiz V. Development of novel phosphorus- containing epoxy resins from renewable resources. J Polym Sci, Part A: Polym Chem 2006;44:6717–27.
  2. Lligadas G, Ronda JC, Galia M, Cadiz V. Synthesis and properties of thermosetting polymers from a phosphorous-containing fatty acid derivative. J Polym Sci, Part A: Polym Chem 2006;44:5630–44.
  3. Ma S, Liu X, Jiang Y, Fan L, Feng J, Zhu J. Synthesis and properties of phosphorus- containing bio-based epoxy resin from itaconic acid. Sci China Chem 2014;57:


  1. Werpy T, Petersen G. In: Top value added chemicals from biomass: volume I– results of screening for potential candidates from sugars and synthesis gas. Golden, CO (US): National Renewable Energy Lab.; 2004.
  2. Zia KM, Noreen A, Zuber M, Tabasum S, Mujahid M. Recent developments and future prospects on bio-based polyesters derived from renewable resources: a review. Int J Biol Macromol 2016;82:1028–40.
  3. Wang Z, Wei T, Xue X, He M, Xue J, Song M, Wu S, Kang H, Zhang L, Jia Q. Synthesis of fully bio-based polyamides with tunable properties by employing itaconic acid. Polymer 2014;55:4846–56.
  4. He M, Wang Z, Wang R, Zhang L, Jia Q. Preparation of bio-based polyamide elastomer by using green plasticizers. Polymers 2016;8:257.
  5. Patil DM, Phalak GA, Mhaske ST. Novel phosphorus-containing epoxy resin from renewable resource for flame-retardant coating applications. J Coat Technol Res


  1. van Krevelen DW. Some basic aspects of flame resistance of polymeric materials.

Polymer 1975;16:615–20.

  1. Baroncini EA, Kumar Yadav S, Palmese GR, Stanzione III JF. Recent advances in bio-based epoxy resins and bio-based epoxy curing agents. J Appl Polym Sci 2016; 133:44103.
  2. Mao W, Li S, Yang X, Cao S, Li M, Huang K, Xia J. Preparation of a flame-retardant epoxy curing agent based on castor oil and study on the curing reaction kinetics.

J Therm Anal Calorim 2017;130:2113–21.

  1. Guo WW, Wang X, Gangireddy CSR, Wang JL, Pan Y, Xing WY, Song L, Hu Y. Cardanol derived benzoxazine in combination with boron-doped graphene toward simultaneously improved toughening and flame retardant epoxy composites.

Compos Part A: Appl S 2019;116:13–23.

  1. Xu X, Wang S, Ma S, Yuan W, Li Q, Feng J, Zhu J. Vanillin-derived phosphorus- containing compounds and ammonium polyphosphate as green fire-resistant systems for epoxy resins with balanced properties. Polym Adv Technol 2019;30:


  1. Yang X, Wang C, Xia J, Mao W, Li S. Study on synthesis of novel phosphorus- containing flame retardant epoxy curing agents from renewable resources and the comprehensive properties of their combined cured products. Prog Org Coat 2017;


  1. Menard R, Negrell C, Fache M, Ferry L, Sonnier R, David G. From a bio-based phosphorus-containing epoxy monomer to fully bio-based flame-retardant thermosets. RSC Adv 2015;5:70856–67.

Copy provided by Alaska Resources Library &

Information Services. Please let us know if you have problems with the document via

Thanks for using ARLIS Interlibrary Loan


word image 2793

Copyright Warning


The copyright law of the United States (Title 17, United States Code) governs the making of photocopies or other reproductions of copyrighted material.


Under certain conditions specified in the law, libraries and archives are authorized to furnish a photocopy or other reproduction. One of these specified conditions is that the photocopy or reproduction is not to be “used for any purpose other than private study, scholarship, or research.” If a user makes a request for, or later uses, a photocopy or reproduction for purposes in excess of “fair use,” that user may be liable for copyright infringement.

word image 2794

Contents lists available at




journal homepage:


word image 718

Chemosphere 227 (2019) 32





Degradation of brominated polymeric flame retardants and effects of generated decomposition products

word image 719

Christoph Koch a, b, *, Bernd Sures a

a Aquatic Ecology and Centre for Water and Environmental Research (ZWU), University Duisburg-Essen, 45141, Essen, Germany b Deutsche Rockwool GmbH & Co. KG, 45966, Gladbeck, Germany

h i g h l i g h t s

The annual production volume of polymeric flame retardants increases constantly.

An important chemical of this group is “Polymeric FR”.

Recent studies have shown that this polymer can degrade into multiple monomers.

So far, the toxic potential of degradation products seems rather limited.

Especially applications outside thermal insulation could be important in the future.

a r t i c l e i n f o

a b s t r a c t

Article history:

Received 24 February 2019

Received in revised form

24 March 2019

Accepted 5 April 2019

Available online 8 April 2019 Handling Editor: Jerzy Falandysz

Keywords: PolyFR






Brominated flame retardants are often associated with adverse environmental effects. Nevertheless, these chemicals are required in order to comply with fire safety standards. Therefore, a better environmental profile is desirable. A “new” class of flame retardants is claimed to fulfil this request while still being feasible for established industrial processes. Different to previous brominated flame retardants, this new group is based on a polymeric structure that could indeed lead to a better environmental profile. However, not much is known about the long-term behaviour of such flame retardants. This short review summarizes what has already been published.

With an annual production volume of 26,000 metric tons, “Polymeric FR” is currently the only industrially produced representative of this group. It has been shown to degrade under specific circumstances (following UV and heat exposure). Detected degradation products cause almost no acute toxicity, whereas chronic toxicity might be relevant. Nevertheless, as long as polymeric flame retardants are only used in building insulation, the actual risk seems to be rather limited.

© 2019 Elsevier Ltd. All rights reserved.

1. Introduction

Flame retardants (FR) are nowadays being used in various products such as electronical equipment, upholstery of furniture, textiles, and thermal insulation (Covaci et al., 2011). The majority of these products consists mainly of plastics and is thus often susceptible to ignition. As a consequence, these products almost always need to contain FRs in order to comply with fire safety requirements (Babrauskas et al., 2012; Cooper et al., 2016). Among all plastic products, applied FRs are mostly halogenated (brominated or chlorinated) organic substances (Birnbaum and Staskal, 2004; de Wit, 2002; Hou et al., 2016; Papachlimitzou et al., 2012; van Mourik et al., 2016). Following the ban and phase out of many so-called “legacy” FRs such as polybrominated biphenyls (PBB) (de Wit, 2002), a tendency to introduce higher halogenated substances to the market can be observed. Examples for these “novel” or “emerging” FRs (Fig. 1) are tetrabromobisphenol A-bis(2,3-dibromopropylether) (TBBPA-DBPE), 2,4,6tris(2,4,6-tribromophenoxy)-1,3,5-triazine (TTBP-TAZ), and 2,2bis(chloromethyl)propane-1,3-diyl-tetrakis (2-chloroethyl)bis

* Corresponding author. Aquatic Ecology and Centre for Water and Environmental Research (ZWU), University Duisburg-Essen, 45141, Essen, Germany. E-mail address: (C. Koch). 0045-6535/© 2019 Elsevier Ltd. All rights reserved.

(phosphate) (commercially known as V6) (Ballesteros-Gomez et al., 2014; Covaci et al., 2011; Stapleton et al., 2011). With an increasing

number of halogens, the molecular mass (and size) of these novel FRs increases as well. For instance, TTBP-TAZ’s average mass is well

word image 720

330 C. Koch, B. Sures / Chemosphere 227 (2019) 329e333

word image 721

Fig. 1. Chemical structure of tetrabromobisphenol A-bis(2,3-dibromopropylether) (TBBPA-DBPE), 2,4,6-tris(2,4,6-tribromophenoxy)-1,3,5-triazine (TTBP-TAZ), 2,2bis(chloromethyl)propane-1,3-diyl-tetrakis (2-chloroethyl)bis (phosphate) (commercially known as V6).

above 1000 Da.

Similar to previous phase-outs, the recent ban of hexabromocyclododecane (HBCD) (Covaci et al., 2006; Koch et al., 2015; Marvin et al., 2011) forced manufacturers of expanded (EPS) and extruded (XPS) polystyrene to find alternative FRs in the area of thermal insulation. However, in contrast to previous substitutions, the developed alternative was not a monomeric compound anymore, but a polymer called “Polymeric FR” (Beach et al., 2017; Jeannerat et al., 2016). With an annual production of 26,000 metric tons, this brominated flame retardant (BFR) is the first polymeric FR (for clarification: “Polymeric FR” e thus the product name in quotes e refers to an individual substance in the group of polymeric flame retardants), which is being used on an industrial scale (ICL, 2015; LANXESS, 2018). It is a block copolymer of polystyrene and brominated polybutadiene and is also known as PolyFR, pFR, Emerald Innovation 3000, FR-122P, and GreenCrest (CAS No. 1195978-93-8, Fig. 2). The only alternative to “Polymeric FR” in PS insulation is a monomeric TBBPA-like alternative which is currently solely used in Japan (Kajiwara et al., 2017) called Milebrome B-972 (CAS No 97416-84-7).

Even though still an additive FR (meaning not covalently bonded to the base material), the manufacturers of “Polymeric FR” claim that this compound is better incorporated into the polystyrene matrix (due to its polymeric structure) and more persistent in the matrix than HBCD. In addition, its high molecular weight of more than 100,000 Da should prevent biological availability and thus results in a better environmental profile (DOW, 2011a; 2011b). These claims were mostly supported in initial assessments of

Fig. 2. Chemical structure of “Polymeric FR” as described by the manufacturer.

environmental protection agencies such as the United States Environmental Protection Agency (EPA, 2014) or the Review Committee of the Stockholm Convention (SC, 2011). Similarly, it was, however, questioned how “Polymeric FR” would behave over a long period of time (US EPA, 2014) leading to the necessity of generating data that allow predictions regarding its environmental fate during its lifetime.

2. Degradation of “Polymeric FR”

Considering that e to our knowledge e “Polymeric FR” is the first industrially produced polymer used as a FR, it is important to understand its long-term behaviour in the environment (including end-of-life scenarios) in order to predict if the utilization of polymers is a sustainable step towards environmentally safe products. One factor for this is its potential stability against biotic and abiotic degradation in the environment. Even though not much is known about the degradation of BFRs in general, it was repeatedly shown that degradation products can have a different e and often higher e toxicity than the original FR (Hill et al., 2018; Martin et al., 2017; Su et al., 2018, 2016, 2014). Even though the degradation of polymeric FRs gained more interest recently, our knowledge is still rather narrow e and mostly limited to molecules with bromine attached to aromatic rings, which is not the case for “Polymeric FR”. Studies for other polymeric BFRs have shown that monomers, which did not react during the production process and therefore contaminate the intended polymer, can become volatile at room temperature (Gouteux et al., 2008) and that temperatures around 700 C can lead to monomeric pyrolysates stemming from thermoplastic resins containing polymeric FRs (Wang, 1999). Even though these studies investigated polymeric FRs, they did not focus on possible degradation of the polymer itself under everyday conditions.

C. Koch, B. Sures / Chemosphere 227 (2019) 329e333 331

Everyday challenges include exposure to UV radiation or heat eabiotic factors known for the degradation of polymers in general (Wypch, 2008). Considering that degradation following simulation of everyday conditions has been observed for similar polymers such as polystyrene (Yousif and Haddad, 2013), it seems possible that “Polymeric FR” has the potential to degrade as well. This process has been the subject of two recently published studies (Koch et al., 2019, 2016). A main focus of the latter studies was on the degradation of “Polymeric FR” following exposure to UV radiation, which is arguably not the most dominant factor for a flame retardant which is only applied in thermal insulation and thus can spend around 50 years behind a façade; shielded against sunlight. Nevertheless, exposure to sunlight and thus UV irradiation can still occur at production (for example after accidental spills), construction or demolition sites (Covaci et al., 2006; Thomsen et al., 2007; Zhang et al., 2013). In addition, UV exposure can also occur if EPS or XPS is not incinerated (Kajiwara et al., 2017; Mark et al., 2015) or chemically recycled (Schlummer et al., 2017), which is foreseen for the future, but brought to landfill. From the experimental perspective, UV treatment is a time-effective procedure to simulate degradation processes (Koch et al., 2016). Therefore, bulk “Polymeric FR” (not incorporated in polystyrene) was irradiated for up to 3 h, which equals a period between 6 h and 9 days (judged on UV-A or UV-B radiation) of natural sunlight during summer on the northern hemisphere (Bais et al., 2011; Koch et al., 2016). Gel Permeation Chromatography spectra indicate a broadening of the molecular mass distribution, most likely via chain scission reaction, that can also lead to crosslinking through radical combination in solid state and is a common mechanism for similar polymers (Koch et al., 2016; Yousif and Haddad, 2013). Accordingly, the polymeric substance was at least partly degraded which leads to several smaller molecules. Degradation of “Polymeric FR” can also be observed via abstraction of bromine, which was suggested based on 1H NMR measurements (Koch et al., 2016). This suggestion is coherent with observations made during the degradation of “Polymeric FR” in solvents. Independent of the type of solvent (rain, reconstituted or distilled water), a loss of approximately 7.5% (mainly not organically bound) of the bromine content of “Polymeric FR” (and thus 4.5% of its total weight) was observed after 3 h of radiation (Koch et al., 2019). During the same time span,1% of the carbon content of “Polymeric FR” was also detected in the solvent. Taking these results into account, it is not surprising that monomeric degradation products were detected via liquid chromatography-mass spectrometry (LC-MS). A number of 75 distinct molecular formulas (with a score above 95) was listed, including about 10% of brominated compounds. Of these, three brominated structures were clearly identified: 2,4,6-tribromo-3hydroxybenzoic acid (THA), 3,5-dibromo-4-hydroxybenzoic acid (DHA) and 5-bromosalicylic acid (BSA). In addition, an isomer of tribromophenol (TBP) was suggested based on the obtained spectra (Koch et al., 2019). Using LC-UV and an artificial radiation period of 35 and 70 h (equalling 6 and 12 months of natural radiation), a comparably lower abundance of photolytic products around 3% by weight was estimated (DOW, 2018).

Experiments have also been conducted with the final product: EPS containing “Polymeric FR”. In this case, an irradiation of 1 h has led to a bromine loss of 40% of what was detected after solely exposing bulk “Polymeric FR” (Koch et al., 2016). The comparably lower loss (and thus most likely lower degree of degradation) was explained by a shielding effect of the surrounding polystyrene against UV radiation. It should be noted, that even though the final insulation product was used during this study, EPS beads with a diameter of approximately 5 mm have been separated beforehand, whereas in reality beads would probably not be separated which would in consequence lead to less degradation. The identification of degradation products of “Polymeric FR” has not yet been attempted with PS foam samples.

An arguably more relevant factor for degradation of BFRs applied in insulation products, is their exposure to heat. Temperatures of above 70 C have been reported during summer in hot attics (Parker, 2005), which affects the installed insulation material (Zirkelbach et al., 2007). Similar temperatures might also occur during the end-of-life of insulation products on landfilling sites (Yesiller et al., 2005; UK EA, 2003). It should be clearly noted, that e even though the degradation via heat exposure was studied e there is currently no information available regarding the reaction to fire, which is likely to give completely different results.

Based on samples that were kept at room temperature in water (over two years) and samples that were exposed to heat (60 C for 36 weeks) without a solvent, it was concluded that “Polymeric FR” (not incorporated in PS foam) does not significantly degrade in either situation (Koch et al., 2019). Instead, only a combination of both factors (60 C for 36 weeks in water) led to a substantial degradation. Following such an exposure, 3% of the carbon content of “Polymeric FR” was detected in the solvent. Contrary, the total bromine concentration as measured via inductively coupled plasma-mass spectrometry (ICP-MS) was much lower than expected (0,0005%), which was explained by evaporation of HBr or Br2 over such a comparably long period. The exposure to heat led to seven identified degradation products, including one containing bromine: 5-bromosalicylic acid (Koch et al., 2019). Considering that these degradation results were only observed in the presence of water, their relevance should be carefully assessed. Similar to the discussion regarding the relevance of UV radiation, water might be a considerable factor during accidental spills, at construction and demolition sites or during the end-of-life of such products. However, water will most often not be present during the actual use phase, i.e. when EPS is installed inside a building. It should also be noted that so far, EPS containing “Polymeric FR” has not been studied following exposure to heat. Incorporated “Polymeric FR” might behave differently under such circumstances.

Different to abiotic degradation, no information is available about biotic degradation of “Polymeric FR”. However, it is a known fact that polymers in general undergo degradation facilitated by microorganisms (Ahmed et al., 2018; Emadian et al., 2017; Iwata, 2015; Laycock et al., 2017). This is also the case for polystyrene (Andrady, 2011), which has structural similarities to “Polymeric FR” and could thus indicate a potential for biotic degradation. In addition, it was shown that abiotic degradation, particularly UV irradiation, can lead to an increased hydrophilicity of “Polymeric FR” (Koch et al., 2016), which is commonly accepted as a factor for easier colonization by microorganisms and subsequent biodegradation (Donlan, 2002; Wang et al., 2012). However, without any biodegradation studies on this specific FR, it is not possible to fully evaluate its potential to undergo biotic degradation.

3. Toxicity of degradation products

The potential toxicity of “Polymeric FR” itself has not yet been the subject of peer-reviewed publications. According to a hazard summary of the United States Environmental Protection Agency (ranging from very low to very high), all evaluated endpoints were classified as low, except the potential for eye irritancy, which was rated as medium (US EPA, 2014). Based on predictive models and/or professional judgement, the persistence of “Polymeric FR” was rated as very high (US EPA, 2014).

Considering the previously described findings regarding the degradation of “Polymeric FR”, one study assessed the aquatic toxicity of some of the identified degradation products using standardized OECD methods (Koch and Sures, 2019). Four individual compounds (THA, DHA, BSA, and TBP) and a mixture of those was tested (which consisted of equal mass concentrations per substance). It should be noted, that the isomer 2,4,6-TBP was used, even though this was not clearly identified as a degradation product. In general, almost no additional ecotoxicological information has been published regarding the identified degradation products. There is only some knowledge available concerning TBP, which indicates a certain potential for aquatic toxicity (Koch and Sures, 2018) although it is also a substance which is naturally synthesised by different groups of marine organisms.

332 C. Koch, B. Sures / Chemosphere 227 (2019) 329e333

The results obtained by Koch and Sures (2019) indicate an almost neglectable acute toxicity in the 48 h daphnia immobilization (no observed effect concentration (NOEC) of 2,5 mg/L per substance for the mixture) and the 72 h algae growth inhibition test (EC50> 10 mg/L). The lowest predicted LC50 (using the QASR OECD Toolbox) for Pimephales promelas after 96 h was calculated around 3 mg/L for an individual degradation product. In addition, a chronic exposure study was performed using the OECD No 211 method, which assesses effects on the reproduction of daphnids. For this test, daphnids were exposed over three weeks to the mixture (maximum concentration of 3 mg/L per substance), which resulted in a NOEC (regarding number of live offspring) of 37 mg/L (Koch and Sures, 2019). The predicted log KOW for the evaluated substances was 4.20 (Koch and Sures, 2019), which indicates a comparably limited potential for bioaccumulation (GHS, 2017).

Considering that this is the only study dealing with the potential toxicity of degradation products of “Polymeric FR” and that it comes with some limitations (limited amount of degradation products, artificial mixture of those and small range of test species), more research is certainly beneficial to asses if polymeric FRs are a suitable alternative for the future.

4. Cause for concern?

Considering that “Polymeric FR” is currently only applied in thermal building insulation, the risk for degradation of this BFR during everyday usage (thus excluding for instance fire) seems comparably low as long as the product is adequately handled at its end-of-life. In this regard and based on the extremely limited amount of studies currently available, it appears that “Polymeric FR” is an important step towards environmentally safer FRs.

It seems however likely for the future, that this or a similar polymeric FR might be used in other areas of application as for instance textiles. The theoretical feasibility of such an application was already shown some years ago with poly (pentabromobenzyl acrylate), which however e based on the limited literature available e gained only little to no commercial and scientific attention (Borms and Georlette et al., 2004). Considering that textiles are commonly exposed to UV radiation, heat, and water (through rain water or during washing), degradation might occur relatively easy as indicated previously. Therefore, it is important to receive a thorough understanding of degradation mechanisms that might be relevant for polymeric FRs and to study potential environmental effects, while always evaluating the relevance of obtained results.


Ahmed, T., Shahid, M., Azeem, F., Rasul, I., Shah, A.A., Noman, M., Hameed, A., Manzoor, N., Manzoor, I., Muhammad, S., 2018. Biodegradation of plastics: current scenario and future prospects for environmental safety. Environ. Sci. Pollut. Res. 25, 7287e7298.

Andrady, A.L., 2011. Microplastics in the marine environment. Mar. Pollut. Bull. 62, 1596e1605.

Babrauskas, V., Lucas, D., Eisenberg, D., Singla, V., Dedeo, M., Blum, A., 2012. Flame retardants in building insulation: a case for re-evaluating building codes. Build. Res. Inf. 40, 738e755.

Bais, A.F., McKenzie, R.L., Bernhard, G., Aucamp, P.J., Ilyas, M., Madronich, S., Tourpali, K., 2011. Ozone depletion and climate change: impacts on UV radiation. Photochem. Photobiol. Sci. 10, 182e198. C0PP90034F.

Ballesteros-Gomez, A., de Boer, J., Leonards, P.E.G., 2014. A novel brominated triazine-based flame retardant (TTBP-TAZ) in plastic consumer products and indoor dust. Environ. Sci. Technol. 48, 4468e4474. es4057032.

Beach, M.W., Hull, J.W., King, B.A., Beulich, I.I., Stobby, B.G., Kram, S.L., Gorman, D.B., 2017. Development of a new class of brominated polymeric flame retardants based on copolymers of styrene and polybutadiene. Polym. Degrad. Stabil. 135, 99e110.

Birnbaum, L.S., Staskal, D.F., 2004. Brominated flame retardants: cause for concern? Environ. Health Perspect. 112, 9e17.

Borms and Georlette 2004, Borms, R., Georlette, P., 2004. Polymere Flammschutzmittel. Kunststoffe 9/2004, pp. 256e260. Available online: https://www.

Cooper, E.M., Kroeger, G., Davis, K., Clark, C.R., Ferguson, P.L., Stapleton, H.M., 2016.

Results from screening polyurethane foam based consumer products for flame retardant chemicals: assessing impacts on the change in the furniture flammability standards. Environ. Sci. Technol. 50, 10653e10660. 10.1021/acs.est.6b01602.

Covaci, A., Gerecke, A.C., Law, R.J., Voorspoels, S., Kohler, M., Heeb, N.V., Leslie, H.A., Allchin, C.R., De Boer, J., 2006. Hexabromocyclododecanes (HBCDs) in the environment and humans: a review. Environ. Sci. Technol. 40, 3679e3688.

Covaci, A., Harrad, S., Abdallah, M.A.E., Ali, N., Law, R.J., Herzke, D., de Wit, C.A., 2011. Novel brominated flame retardants: a review of their analysis, environmental fate and behaviour. Environ. Int. 37, 532e556.


de Wit, C.A., 2002. An overview of brominated flame retardants in the environment. Chemosphere 46, 583e624.

Donlan, R.M., 2002. Biofilms: Microbial life on surfaces. Emerg. Infect. Dis. 8, 881e890.

DOW 2011a, Beulich, I., 2011. New Class of Brominated Polymeric Flame Retardants. Fire Resistance in Plastics (Cologne, Germany).

DOW, 2011b, Davis, J.W., 2011. Strategic Approach towards More Environmentally Sustainable Flame Retardants (Midland, USA).

DOW 2018, Davis, J.W., Hu, J., Kearns, K.L., Beach, M., Kram, S., Lai, Y., Stutzman, J., 2018. Environmental fate testing of a styrene-butadiene brominated copolymer flame retardant. In: Poster for the Society of Environmental Toxicology and Chemistry (SETAC) Conference.

Emadian, S.M., Onay, T.T., Demirel, B., 2017. Biodegradation of bioplastics in natural environments. Waste Manag. 59, 526e536. 2016.10.006.

GHS 2017, 2017. United Nations (UN): Globally Harmonized System of Classification and Labelling of Chemicals (GHS), Part 4 e Environmental Hazards. Revision 7. Available online: ghs/ghs_rev07/English/04e_part4.pdf. (Accessed 17 March 2018).

Gouteux, B., Alaee, M., Mabury, S.A., Pacepavicius, G., Muir, D.C.G., 2008. Polymeric brominated flame retardants: are they a relevant source of emerging brominated aromatic compounds in the environment? Environ. Sci. Technol. 42, 9039e9044.

Hill, K.L., Mortensen, Å.-K., Teclechiel, D., Willmore, W.G., Sylte, I., Jenssen, B.M., Letcher, R.J., 2018. In vitro and in silico competitive binding of brominated polyphenyl ether contaminants with human and gull thyroid hormone transport proteins. Environ. Sci. Technol. 52, 1533e1541. acs.est.7b04617.

Hou, R., Xu, Y., Wang, Z., 2016. Review of OPFRs in animals and humans: absorption, bioaccumulation, metabolism, and internal exposure research. Chemosphere 153, 78e90.

ICL 2015, November 2015. ICL-IP: ICL-IP Develops New Grades of Polymeric Flame Retardant for EPS and XPS. Published in Additives for Polymers.

Iwata, T., 2015. Biodegradable and bio-based polymers: future prospects of ecofriendly plastics. Angew. Chem. Int. Ed. 54, 3210e3215. 10.1002/anie.201410770.

Jeannerat, D., Pupier, M., Schweizer, S., Mitrev, Y.N., Favreau, P., Kohler, M., 2016. Discrimination of hexabromocyclododecane from new polymeric brominated flame retardant in polystyrene foam by nuclear magnetic resonance. Chemosphere 144, 1391e1397.

Kajiwara, N., Noma, Y., Sakai, S., 2017. Environmentally sound destruction of hexabromocyclododecanes in polystyrene insulation foam at commercial-scale industrial waste incineration plants. J. Environ. Chem. Eng. 5, 3572e3580.

Koch, C., Sures, B., 2018. Environmental concentrations and toxicology of 2,4,6tribromophenol (TBP). Environ. Pollut. 233, 706e713. j.envpol.2017.10.127.

Koch, C., Sures, B., 2019. Ecotoxicological characterization of possible degradation products of the polymeric flame retardant “Polymeric FR” using algae and

Daphnia OECD tests. Sci. Total Environ. 656, 101e107. j.scitotenv.2018.11.207.

Koch, C., Schmidt-Kotters,€ T., Rupp, R., Sures, B., 2015. Review of hexabromocyclododecane (HBCD) with a focus on legislation and recent publications concerning toxicokinetics and -dynamics. Environ. Pollut. 199, 26e34.

Koch, C., Dundua, A., Aragon-Gomez, J., Nachev, M., Stephan, S., Willach, S., Ulbricht, M., Schmitz, O.J., Schmidt, T.C., Sures, B., 2016. Degradation of polymeric brominated flame retardants: development of an analytical approach using PolyFR and UV irradiation. Environ. Sci. Technol. 50, 12912e12920.

Koch, C., Nachev, M., Klein, J., Koster, D., Schmitz, O.J., Schmidt, T.C., Sures, B., 2019.€ Degradation of the polymeric brominated flame retardant “polymeric FR” by heat and UV exposure. Environ. Sci. Technol. 53, 1453e1462. 10.1021/acs.est.8b03872.

LANXESS 2018, 2018. LANXESS: Feuersicherheit e Zeit zu entkommen. Published in


Laycock, B., Nikolic, M., Colwell, J.M., Gauthier, E., Halley, P., Bottle, S., George, G., 2017. Lifetime prediction of biodegradable polymers. Prog. Polym. Sci. 71, 144e189.

Mark, F.E., Vehlow, J., Dresch, H., Dima, B., Grüttner, W., Horn, J., 2015. Destruction of the flame retardant hexabromocyclododecane in a full-scale municipal solid waste incinerator. Waste Manag. Res. 33, 165e174. 0734242X14565226.

C. Koch, B. Sures / Chemosphere 227 (2019) 329e333 333

Martin, O.V., Evans, R.M., Faust, M., Kortenkamp, A., 2017. A human mixture risk assessment for neurodevelopmental toxicity associated with polybrominated diphenyl ethers used as flame retardants. Environ. Health Perspect. 125, 1e11.

Marvin, C.H., Tomy, G.T., Armitage, J.M., Arnot, J.A., McCarty, L., Covaci, A., Palace, V.P., 2011. Hexabromocyclododecane: current understanding of chemistry, environmental fate and toxicology and implications for global management. Environ. Sci. Technol. 45, 8613e8623.

Papachlimitzou, A., Barber, J.L., Losada, S., Bersuder, P.R.C., Law, R.J., 2012. A review of the analysis of novel brominated flame retardants. J. Chromatogr. A 1219, 15e28.

Parker 2005, 2005. Parker, D for the Florida Solar Energy Center/University of Central Florida. Literature Review of the Impact and Need for Attic Ventilation in Florida Homes. Available online: pdf/FSEC-CR-1496-05.pdf. (Accessed 11 March 2018).

SC 2011, 2011. Stockholm Convention on Persistent Organic Pollutants: Report of the Persistent Organic Pollutants Review Committee on the Work of its Seventh

Meeting. Addendum: Risk management evaluation on hexabromocyclododecane. Available online: POPsReviewCommittee/ReportsandDecisions/tabid/3309/Default.aspx.

(Accessed 10 February 2014).

Schlummer, M., M€aurer, A., Wagner, S., Berrang, A., Fell, T., Knappich, F., 2017. Recycling of flame retarded waste polystyrene foams (EPS and XPS) to PS granules free of hexabromocyclododecane (HBCDD). Adv. Recycl. Waste Manag. 1e6.

Stapleton, H.M., Klosterhaus, S., Keller, A., Ferguson, P.L., Van Bergen, S., Cooper, E., Webster, T.F., Blum, A., 2011. Identification of flame retardants in polyurethane foam collected from baby products. Environ. Sci. Technol. 45, 5323e5331.

Su, G., Letcher, R.J., Crump, D., Farmahin, R., Giesy, J.P., Kennedy, S.W., 2014. Photolytic degradation products of two highly brominated flame retardants cause cytotoxicity and mRNA expression alterations in chicken embryonic hepatocytes. Environ. Sci. Technol. 48, 12039e12046. es503399r.

Su, G., Letcher, R.J., Crump, D., Farmahin, R., Giesy, J.P., Kennedy, S.W., 2016. Sunlight irradiation of highly brominated polyphenyl ethers generates polybenzofuran products that alter dioxin-responsive mRNA expression in chicken hepatocytes. Environ. Sci. Technol. 50, 2318e2327.

Su, G., Letcher, R.J., Farmahin, R., Crump, D., 2018. Photolysis of highly brominated flame retardants leads to time-dependent dioxin-responsive mRNA expression in chicken embryonic hepatocytes. Chemosphere 194, 352e359. https://

Thomsen, C., Molander, P., Daae, H.L., Janak, K., Frøshaug, M., Liane, V.H., Thorud, S.,

Becher, G., Dybing, E., 2007. Occupational exposure to hexabromocyclododecane at an industrial plant. Environ. Sci. Technol. 41, 5210e5216.

UK EA 2003) UK Environment Agency, 2003. Guidance on Monitoring of Landfill Leachate, Groundwater and Surface Water. Available online:

uk/government/uploads/system/uploads/attachment_data/file/321602/ LFTGN02.pdf. (Accessed 26 March 2018).

US EPA 2014, 6 2014. United States Environmental Protection Agency: Flame Retardant Alternatives for Hexabromocyclododecane (HBCD) e Final Report, 06/2014. Available online: documents/hbcd_report.pdf. (Accessed 11 January 2016).

van Mourik, L.M., Gaus, C., Leonards, P.E.G., de Boer, J., 2016. Chlorinated paraffins in the environment: a review on their production, fate, levels and trends between

2010 and 2015. Chemosphere 155, 415e428.


Wang, F.C.-Y., 1999. Analysis of polymeric flame retardants in engineering thermoplastics by pyrolysis gas chromatography with an atomic emission detector. Anal. Chem. 71, 2037e2045.

Wang, R., Neoh, K.G., Shi, Z., Kang, E.T., Tambyah, P.A., Chiong, E., 2012. Inhibition of escherichia coli and proteus mirabilis adhesion and biofilm formation on medical grade silicone surface. Biotechnol. Bioeng. 109, 336e345. https://

Wypch 2008, Wypch, G., 2008. Handbook of Material Weathering. ChemTec Publishing, Toronto, Ontario.

Yesiller, N., Hanson, J.L., Liu, W.-L., 2005. Heat generation in municipal solid waste landfills. J. Geotech. Geoenviron. Eng. 131, 1330e1344. (ASCE)1090-0241 (2005)131:11(1330).

Yousif, E., Haddad, R., 2013. Photodegradation and photostabilization of polymers, especially polystyrene: review. SpringerPlus 2, 398. 2193-1801-2-398.

Zhang, Y., Ruan, Y., Sun, H., Zhao, L., Gan, Z., 2013. Hexabromocyclododecanes in surface sediments and a sediment core from Rivers and Harbor in the northern Chinese city of Tianjin. Chemosphere 90, 1610e1616. j.chemosphere.2012.08.037.

Zirkelbach et al., 2007, Zirkelbach, D., Künzel, H.M., Bludau, C., 2007. (Fraunhofer IBP): Durability Assessment of Glass Fibre Insulation in Flat Roof Constructions. Available online: documents/Publikationen/oeffentliche_Berichte/Englisch/oeVB_eng_1_tcm4530993.pdf. (Accessed 18 May 2018).

Order a unique copy of this paper
(550 words)

Approximate price: $22