Polymer Degradation and Stability 130 (2016) 173e181

Contents lists available at ScienceDirect

Polymer Degradation and Stability

journal homepage: www.elsevier.com/locate/polydegstab

Synergistic flame-retardant behavior and mechanisms of aluminum

poly-hexamethylenephosphinate and phosphaphenanthrene in epoxy
resin
Jingyu Wang, Lijun Qian*, Zhigang Huang, Youyou Fang, Yong Qiu
School of Materials Science & Mechanical Engineering, Beijing Technology and Business University, Beijing 100048, PR China

a r t i c l e i n f o a b s t r a c t

Article history: The flame retardants aluminum poly-hexamethylenephosphinate (APHP) and 9,10-dihydro-9-oxa-10-
Received 29 April 2016 phosphaphenanthrene 10-oxide(DOPO) were incorporated into diglycidyl ether of bisphenol A (EP)
Received in revised form thermoset, and then the synergistic flame-retardant behavior and mechanism of APHP/DOPO were
6 June 2016
investigated. Comparing with the thermosets with 6%APHP and 6%DOPO alone, 2%APHP/4%DOPO/EP
Accepted 13 June 2016
thermosets obtained the higher limited oxygen index, higher UL94 rating, decreased peak of heat release
Available online 15 June 2016
rate and less total heat release from cone calorimeter test. The results reveal a synergistic effect between
APHP and DOPO. The synergistic effect of APHP/DOPO in gaseous phase obviously reduced effective heat
Keywords:
Flame retardant
of combustion, which implies the better flame inhibition effect through quenching free radical chain
Phosphinate reaction of combustion. The synergistic charring effect in condensed phase led to the higher char yield,
DOPO which locked more carbonaceous contents in residue and form more barrier to heat spreading. All the
Synergistic effect results were caused by the early decomposed DOPO that interacted with the later decomposed APHP to
Epoxy resin produce more char and decease release of the inflammable gas. Therefore, the burning intensity of APHP/
DOPO thermosets obviously was weakened by the synergistic effect of APHP/DOPO.
© 2016 Elsevier Ltd. All rights reserved.

1. Introduction widely used to react with epoxy resin to prepare flame-retardant

In the past several decades, flame-retardant epoxy resins have
become important advanced materials in electronic and electrical
equipment industries due to their excellent adhering, physical-
mechanical, electric, and flame-retardant properties [1e6]. There-
fore, they are widely applied as binders in printed circuit boards
and as packaging materials in light-emitting diode illuminators
[7e12]. To enhance flame-retardant efficiency and comprehensive
properties of epoxy resins, a series of novel flame retardants
especially for phosphorus-based flame retardants were prepared
and incorporated into epoxy resins because of their halogen-free
and high flame retardancy properties [13e19]. Among these
flame retardants, some reactive and additive phosphorus flame
retardants are the most important ones [20e29].
As an important reactive phosphorus flame retardant, 9,10-
dihydro-9-oxa-10-phosphaphenanthrene 10-oxide(DOPO) was
* Corresponding author. Gengyun Building No.516, Fucheng Road No.33, Haidian
District, Beijing, PR China.
E-mail address: qianlj@th.btbu.edu.cn (L. Qian).
thermosets and also introduced into curing agent to obtain
flame-retardant curing system [20,30,31]. Recently, most of the
studies about DOPO have focused on the novel additives con-
structed by phosphaphenanthrene and several functional groups
such as cyclotriphosphazene [14,32], silsesquioxane [33], triazene
[16,34,35], and other groups [36e38]. These additives can endow
epoxy resin thermosets by high flame retardancy, and confirm
flame-retardant group synergistic effect between DOPO and other
functional groups by constructing novel molecules. Though the
studies on constructing DOPO-based additives have made progress
on developing novel and various high efficiency flame retardants, it
is still essential to explore the flame-retardant behavior of com-
posites because it has access to seek for flame-retardant epoxy
resin thermosets with higher performance [23,39e43].
Recently, a novel alkyl-phosphinate flame retardant, aluminum
poly-hexamethylenephosphinate (APHP, Scheme 1), was prepared
in our laboratory [44]. In former study, APHP can endow diglycidyl
ether of bisphenol A (EP) thermosets with better flame retardancy
in both condensed and gas phase. But the APHP/EP thermosets still
failed to V-0 rating. In this study, we applied APHP into a diglycidyl

http://dx.doi.org/10.1016/j.polymdegradstab.2016.06.010
0141-3910/© 2016 Elsevier Ltd. All rights reserved.
174 J. Wang et al. / Polymer Degradation and Stability 130 (2016) 173e181
Scheme 1. Molecular structures of APHP.
ether of bisphenol A (DGEBA) pre-reacted with DODO and then
cured with 4,40 -diamino-diphenyl methane (DDM). After the
thermosets were investigated, a rising flame retardancy and
significantly synergistic effect between APHP and DOPO were ob-
tained in the flame-retardant epoxy resin thermosets. In this study,
quantitative assessment on the synergistic effect of APHP and
DOPO also was carried out [45].
2. Experimental
2.1. Material
Aluminum poly-hexamethylenephosphinate(APHP) was pre-
pared in our laboratory. 9,10-Dihydro-9-oxa-10-
phosphaphenanthrene-10-oxide (DOPO) was supplied by
Shanghai Eutec Chemical Co., China. The molecular structures of
APHP were shown in Scheme 1. The diglycidyl ether of bisphenol A
(EP, commercial name: E-51), was purchased from Blue Star New
Chemical Material Co. Ltd., China. 4,40 -Diamino-diphenyl methane
(DDM) was purchased from Sinopharm Chemical Reagent Co. Ltd.,
China.
2.2. Preparation of flame-retardant EP thermosets and the control
samples
Typically, the 2%APHP/4%DOPO/EP thermoset sample with 2 wt
% APHP and 4 wt% DOPO was prepared via the method: DOPO
(5.28 g) was melted in DGEBA (100 g) at 150  C, and then PeH bond
in DOPO reacted with the epoxy group from DGEBA in ring opening
reaction for 5 h. The reaction formula is as shown in Scheme 2. Then
APHP (2.64 g) was incorporated into the mixture. After APHP was
evenly dispersed in DGEBA, the equivalent mole DDM (24.07 g) of
the reserved epoxy groups (decreased by the mole of addition
DOPO) was added into the mixture at 110  C and blended thor-
oughly. After the mixture was degassed at 120  C for 3 min, it was
Scheme 2. Ring opening reaction between DGEBA and DOPO.
poured into preheated mold and cured at 120  C for 2 h and then at
170  C for 4 h. Then, 2%APHP/4%DOPO/EP thermoset sample was
obtained. Other samples were prepared using the same method
after changing their compositions according to Table 1.
The samples were labeled based on the mass fraction of APHP
and DOPO in the epoxy resin thermoset. The thermosets with 6%
APHP and 6%DOPO alone was labeled as 6%APHP/EP and 6%DOPO.
The control sample, DGEBA was cured to prepare thermoset in the
same manner but without the addition of the flame retardant APHP
and DOPO, and the sample was named as pure EP. The formulations
of DGEBA, DDM, APHP and DOPO in each epoxy resin thermosets
are listed in Table 1.
2.3. Characterizations
Thermogravimetric analysis (TGA) was performed using a TA
instrument Q5000 IR thermal gravimetric analyzer. The sample was
placed in an alumina crucible and heated from 50  C to 700  C at
the rate of 20  C/min in N2 atmosphere.
The limited oxygen index (LOI) value was measured using an FTT
(Fire Testing Technology, UK) Dynisco LOI instrument according to
ASTM D2863-97 (sample dimension: 130 mm  6.5 mm  3.2 mm).
The LOI measurement for each specimen was repeated three times,
and their error values were ±0.5%. The vertical burning test for the
UL 94 combustion level was performed according to ASTM D 3801
(sample dimension: 125 mm  12.7 mm  3.2 mm). Fire behavior
was characterized using FTT cone calorimeter according to ISO5660
at an external heat flux of 50 kW/m2 (sample dimension:
100 mm  100 mm  3 mm). The measurement for each specimen
was repeated two times, and the error values of the typical cone
calorimeter data were reproducible within ±10%. All the data were
collected during combustion.
The micromorphology images of the residues after cone calo-
rimeter test were obtained using a FEI Quanta 250 FEG field-
emission scanning electron microscope at high vacuum condi-
tions with a voltage of 30 kV. The element contents of residues from
cone calorimeter test were investigated via an AMETEK Quanta 250
FEG/EDS Energy Dispersive Spectrometer. The tested specimens
were obtained from the mixture of residues with sufficiently mixed
and grinded, and the results were the average of the three times
repeated tests which were all reproducible within ±5%.
To recognize the pyrolysis fragments of APHP/DOPO/EP ther-
mosets, a Shimadzu GC-MS-QP5050A gas chromatography-mass
spectrometer equipped with a PYR-4A pyrolyzer was employed.
The helium (He) was utilized as carrier gas for the volatile products.
The injector temperature was 250  C, the temperature of GC/MS
interface was 280  C and the cracker temperature was 550  C.
3. Results and discussion
3.1. Fire behavior: LOI and UL94 test
The flame-retardant properties of all the thermosets were
determined using LOI and UL94 vertical burning tests. And the
 J. Wang et al. / Polymer Degradation and Stability 130 (2016) 173e181 175

Table 1
Formulations of flame-retardant EP thermosets and pure EP.

Samples DGEBA (g) DDM (g) APHP DOPO P-content

(g) (wt%) (g) (wt%) (wt%)

1%APHP/1%DOPO/EP 100.00 24.99 1.28 1.00 1.28 1.00 0.35

2%APHP/2%DOPO/EP 100.00 24.68 2.60 2.00 2.60 2.00 0.69
3%APHP/3%DOPO/EP 100.00 24.37 3.97 3.00 3.97 3.00 1.04
1%APHP/2%DOPO/EP 100.00 24.69 1.29 1.00 2.57 2.00 0.49
2%APHP/4%DOPO/EP 100.00 24.07 2.64 2.00 5.28 4.00 0.98

Pure EP 100.00 25.28 e e e e e

6%APHP/EP 100.00 25.28 8.00 6.00 e e 1.22
6%DOPO/EP 100.00 23.52 e e 7.88 6.00 0.86

corresponding results are presented in Table 2. The LOI value of 6%

APHP/EP and 6%DOPO/EP samples increased to 33.1 vol % and
39.0 vol % respectively, and the LOI value of the pure EP sample was
only 26.0 vol %. When APHP and DOPO (3%/3% or 2%/4% weight in
thermosets) were incorporated into the thermosets, the LOI value
of 3%APHP/3%DOPO/EP and 2%APHP/4%DOPO/EP thermosets
further increased to 39.3 vol % and 39.5 vol %. Further, in UL94 test,
the thermoset with 6%APHP can avoid the dripping but still cannot
be rated, and that with 6%DOPO just passed UL94 V-1 rating. But
when APHP/DOPO with certain ratio were simultaneously incor-
porated into thermosets, different results can be obtained. 2%APHP/
4%DOPO/EP thermosets can pass V-0 rating with the same total
loadings as that of APHP or DOPO in thermosets. 3%APHP/3%DOPO/
EP still can only be rated V-1 and failed to pass V-0 rating.
Because APHP/DOPO 2%APHP/4%DOPO/EP sample exhibited the
higher LOI value and the higher flame retardant rating than the
thermosets with same amount of DOPO or APHP did. It revealed
that APHP and DOPO in a suitable ratio would bring the synergistic
Fig. 1. HRR curves of thermosets by cone calorimeter test with 50 kW/m2.
flame retardant effect to epoxy thermosets, and thus imposing the
better flame-retardant properties to epoxy thermosets.

3.2. Fire behavior: cone calorimeter test

Cone calorimeter test was conducted to investigate sufficiently

the synergistic flame-retardant behaviors of APHP and DOPO in the
EP thermosets. The curves of heat release rate (HRR) and the total
heat release are shown in Figs. 1 and 2, respectively., The main
characteristic parameters, such as the time to ignition (TTI), total
heat release (THR), peak of heat release rate (pk-HRR), residue yield
(Residue), total mass loss (TML), average value of effective heat of
combustion (av-EHC), total smoke release (TSR), average value of
specific extinction area (av-SEA), average CO yield (av-COY) and
average CO2 yield (av-CO2Y), are summarized in Table 3.
As shown in Table 3, comparing with the pure EP thermosets,
the TTIs of the 6%APHP/EP thermosets were hardly changed with

Table 2
LOI value and UL94 rating of EP thermosets.

Samples LOI (vol%) UL94

Fig. 2. The THR curves of thermosets by cone calorimeter test with 50 kW/m2.
Rating Dripping

1%APHP/1%DOPO/EP 31.7 Unrated No

2%APHP/2%DOPO/EP 37.2 V-1 No
3%APHP/3%DOPO/EP 39.3 V-1 No the incorporation of APHP, while that of 6%DOPO/EP thermosets
1%APHP/2%DOPO/EP 34.7 V-1 No was obviously shortened to 44s. It indicated that the decrease of
2%APHP/4%DOPO/EP 39.5 V-0 No
TTIs can be attributed to the addition of DOPO because the TTIs of
6% APHP/EP 33.1 Unrated No thermosets with both APHP and DOPO also gradually decreased
6% DOPO/EP 38.5 V-1 No with increasing mass fraction of DOPO in flame retardants. The
EP 26.2 Unrated Yes
addition of DOPO can make thermosets decompose in advance
The bold values given is the optimal sample with the best flame retardant perfor- while APHP do not so. With the development of the thermal
mance, supporting the main opinion of component synergistic effect.
176 J. Wang et al. / Polymer Degradation and Stability 130 (2016) 173e181
Table 3
Typical parameters from cone calorimeter test of thermosets.
Samples TTI (s) pk-HRR (kW/m2)
2%APHP/4%DOPO/EP 46 510
3%APHP/3%DOPO/EP 50 539
6%APHP/EP 55 603
6%DOPO/EP 44 725
EP 56 1420
Samples av-EHC (MJ/kg) TSR (m2/m2)
2%APHP/4%DOPO/EP 17.4 4692
3%APHP/3%DOPO/EP 17.5 4511
6%APHP/EP 20.9 4579
6%DOPO/EP 19.0 5106
EP 29.6 5008
The bold values given is the optimal sample with the best flame retardant performance, supporting the main opinion of component synergistic effect.
decomposition process and the decomposition temperature rising,
APHP reached its decomposition temperature and began to work
with DOPO together rapidly.
The HRR curves are shown in Fig. 1. Comparing with the pk-HRR
values of flame-retardant thermosets with APHP or DOPO, the
thermosets with equivalent APHP/DOPO mixture showed the less
pk-HRR values. Also, the THR values of 3%APHP/3%DOPO/EP and 2%
APHP/4%DOPO/EP exhibit as similar results as pk-HRR values. The
results reveal that not only the combustion intensity but also the
total combustion reaction quantity of thermosets had been
inhibited more effectively when the two components were applied
in epoxy thermosets together. The phenomena preliminarily
disclose the flame retardant synergistic effect between the two
components. Moreover, the THR curves in Fig. 2 also disclosed
another phenomenon that the APHP/DOPO/EP thermosets had the
earlier heat release at the onset decomposition stage but ulti-
mately they produced the less heat comparing with the thermo-
sets only containing the single flame retardant. During
combustion, we deduced that DOPO group would decompose early
and promote the matrix decomposing, and then interacted with
APHP, which bring benefits to reduce the heat release and com-
bustion intensity [32,35]. The flame retardant action mode will be
analyzed in detail in the subsequent discussion. The flame retar-
dant actions in gas phase and condensed one will be investigated
respectively.
In condensed phase, the TML and char yield parameters
exhibited the action effect of DOPO/APHP. TML, represents the
total mass loss which is opposite to the residue yield in cone
calorimeter test. And the residue yield can indicate the charring
effect during combustion in cone calorimeter. In contrast to pure
EP sample, the flame-retardant thermosets contain APHP and
DOPO alone reduced the TML values to 78.9% and 83.7%. But
adding APHP/DOPO into thermosets, the TML values of 3%APHP/3%
DOPO/EP and 2%APHP/4%DOPO/EP thermosets all further
decreased to 72.1% and 76.2%, respectively. The less matrix were
decomposed and the more char was reserved in the APHP/DOPO/
EP system comparing with 6%APHP/EP and 6%DOPO/EP thermo-
sets. Therefore, it can be confirmed the synergistic effect between
DOPO and APHP in promoting formation of more additional
carbonaceous char. The results imply that the APHP/DOPO system
can impose the better flame-retardant effect in condensed phase
to thermosets because the less decomposed matrix reduced the
fuels and the more char inhibited the heat spreading to inner
matrix during combustion. During matrix combustion, we
deduced that DOPO group linked in matrix should decompose
early and release acid to promote matrix charring, and then begin
to co-work with APHP in flame retardant matrix. The DOPO group
and APHP commonly worked to lock more EP matrix in residue,
THR (MJ/m2) TML (wt%) Residue (wt%)
58 76.2 23.8
63 72.1 27.9
69 78.9 21.1
70 83.7 12.3
116 92.2 7.8
av-SEA (m2/kg) av-COY (kg/kg) av-CO2Y (kg/kg)
1399 0.10 1.26
1252 0.09 1.25
1386 0.09 1.58
1389 0.11 1.42
1033 0.08 1.88
thereby resulting in high char yield.
The effect of DOPO/APHP system in gaseous phase was also
discussed after it was confirmed that the system brought
enhancement of flame retardant effect in condensed phase. In
gaseous phase, EHC is particularly pivotal parameter to evaluate the
flame retardant effect. The av-EHC value discloses the burning
degree of volatiles in gas phase during combustion. In Table 3, the
av-EHC of 6%DOPO/EP was lower than that of 6%APHP/EP, which
indicated that stronger gas-phase flame-retardant effect of DOPO.
When DOPO and APHP were commonly incorporated into matrix,
the av-EHC value should locate between the values of DOPO/EP and
APHP/EP. But the results showed that the less av-EHC values were
obtained than DOPO/EP and APHP/EP. Therefore, it reveals that the
av-EHC values of DOPO/APHP/EP obtained more action from syn-
ergistic effect of DOPO and APHP. The synergistic effect of DOPO/
APHP endowed the samples with the effective action on quenching
free radicals of chain reaction of combustion, thereby reducing the
burning ratio of fuels.
The other parameters, such as TSR, av-SEA, av-COY and av-CO2Y
in cone test, provided both the combustion state and secondary fire
hazard effect. The TSR values of flame retardant samples all were
reduced due to more matrix charring. It implies that the more
smoke particles were locked or not formed during combustion,
which should be caused by the two flame retardants. The less
smoke release contributes to decreasing fire hazard. When the
smoke data were investigated by another way, the extra informa-
tion was obtained. The av-SEA represents the smoke values pro-
duced by unit mass materials. The 2%APHP/4%DOPO/EP sample also
sustained the similar av-SEA values as those of 6%APHP/EP and 6%
DOPO/EP, which testified the joint flame inhibition effect of the two
flame retardants because the quenching effect from flame re-
tardants will lead to the incomplete combustion and formation of
more smoke.
Moreover, the obvious reduction in av-EHC of DOPO/APHP/EP
can explain the changes in both COY and CO2Y. As shown in Table 3,
in contrast to the pure EP thermosets, the av-COY of each DOPO/
APHP/EP thermosets did not change obviously while the av-CO2Y
decreased obviously. But the ratios of COY in contrast to CO2Y
obviously increased, which implies that more fuels burn incom-
pletely. The results reveal the flame inhibiting effect on volatiles
during combustion, thus resulting in more incomplete combustion
products (CO) and less complete combustion products (CO2).
According to the above results, DOPO and APHP can bring the
better flame retardant effect to thermosets than anyone of the two
flame retardants in both condensed phase and gaseous phase. The
flame retardant synergistic effect is demonstrated by the discus-
sion. Especially for 2%APHP/4% DOPO thermosets, the appropriate
mass fraction of APHP and DOPO can endow the thermosets with
 J. Wang et al. / Polymer Degradation and Stability 130 (2016) 173e181
the better performance in both LOI test, UL 94 and cone calorimeter
test.
3.3. Fire behavior: quantitative assessment of flame retardant
synergistic effect
For explaining the flame retardant synergistic effect more
directly, the quantitative assessment by analyzing the results from
cone calorimeter was carried out [45]. According to the evaluation
method proposed by Schartel B, the flame retardant effects on
combustion can be divided to three parts: (1) flame inhibition
effect; (2) the charring effect (decrease fuel); (3) barrier and
protective effect from char layer. The first effect is in gaseous
phase, and the last two effects are in condensed phase. All the
effects can be quantitatively calculated their change. Firstly, EHC
which can be regarded as the heat release by unit mass fuel, is the
ratio of HRR to mass loss rate measured at a certain point.
Therefore, it is clear that EHC quantitatively show the flame in-
hibition effect in Formula (1). The charring effect also can be
quantitatively obtained from the residue weight in Formula (2).
The decreased ratio of pk-HRRFR-EP of each samples were much
smaller than that of THRFR-EP. The additional reduction was caused
by the barrier and protective effect of the intumescent char.
Therefore, the calculating method of barrier and protective effect
is shown in Formula (3) [45].
Flame inhibition effect ¼ 1  EHCFREP/EHCEP
Charring effect ¼ 1  TMLFREP/TMLEP
Barrier and protective effect ¼ 1(pk-HRRFREP/pk-HRREP)/(THRFREP/
THREP) (3)
According to the Formula (1e3), the three different flame
retardant effects can be quantitative assessed and the results were
listed in Table 4, which can reveal the action route and quantity of
APHP/DOPO synergistic effect.
As shown in Table 4, the flame inhibition effect of 2%APHP/4%
DOPO/EP system increased to 41.2%, which has outstanding rise
comparing with 29.4% of 6%APHP/EP and 36.8% of 6%DOPO/EP. The
results confirmed the increased synergistic effect of APHP/DOPO in
gaseous phase. The charring effect of 2%APHP/4%DOPO/EP also was
enhanced to 17.4%, which is higher than 14.4% of 6%APHP/EP and
9.2% of 6%DOPO/EP. It implies the stronger charring ability was
obtained from the synergistic effect of APHP/DOPO in condensed
phase. The barrier and protective effect of 2%APHP/4%DOPO/EP
(28.2%) is also better than 15.4% of 6%DOPO/EP but is little worse
than 28.6% of 6%APHP. The barrier and protective effect of APHP/
DOPO locates between APHP and DOPO, but the results also show
synergistic effect if we calculate the synergistic effect according to
the corresponding ratio of APHP and DOPO in thermosets. In sum,
the synergistic flame retardant effects of APHP/DOPO in flame in-
hibition, charring effects and barrier and protective effect all exert
their actions to increase flame retardancy of thermosets. Therefore,
APHP/DOPO system can produce an excellent flame retardant
Table 4
Quantitative assessment of APHP/DOPO synergistic flame retardant effect.
Samples Flame inhibition effect (%)
2%APHP/4%DOPO/EP 41.2
3%APHP/3%DOPO/EP 40.8
6%APHP/EP 29.4
6%DOPO/EP 36.8
177
action in both gaseous phase and condensed phase through their
synergistic effect.
3.4. Morphology and element compositions of the residues from
cone calorimeter test
Fig. 3 shows the macroscopic digital images of the APHP/EP
residues from cone calorimeter test. The residue from the pure EP
sample left less amounts in a badly broken status with few amount
holes, which implies the worse charring ability of pure EP itself. The
residual char yields and expansion ratios of the other samples were
elevated after the incorporation of APHP and DOPO in thermosets.
Particularly, the synergistic one, 2%APHP/4%DOPO/EP thermosets
possessed higher char yields and obviously well expanded residual
char morphology. Thus, the synergistic effect between APHP and
DOPO should not only increase the quantities of the residue but also
elevate the expansion ratios of intumescent layer and make the
char layer more stabilization, thus resulting in better flame retar-
dant effect in condensed phase.
To explore the flame-retardant charring mechanism further,
SEM analysis was also conducted. Fig. 4 typically shows the residue
SEM images of 2%APHP/4%DOPO/EP, 6%APHP/EP, 6%DOPO/EP and
pure EP samples from cone calorimeter test. The residue of pure EP
sample shows many open holes with different sizes in Fig. 4d. The
open holes in the residue of pure EP sample provided channels for
the combustible volatiles generated from the inner matrix to gas
(1) phase, and accordingly increasing the combustion intensity. The
residue of 6%APHP/EP sample in Fig. 4b presents a rigid surface
(2) with a few obvious raised structure and stretched tracks, which
indicates the toughness enhancement of the thermoset melt during
combustion. The residue of 6% DOPO/EP sample in Fig. 4c presents
broken bubble structure, and some undamaged cell instead of the
open holes of pure EP samples. The improvement in sealing vola-
tiles can be seen in both 6%APHP/EP and 6%DOPO/EP samples.
Fig. 4a, the whole morphology of the synergistic sample 2%APHP/
4%DOPO/EP shows some bubbles with thin film, which revealed
that the residue melt is tough. The tough melt state contributed to
higher expanding ratio of residue and sealing more fuels, which
will led to decreasing the combustion intensity in condensed phase.
According to the former discussion in cone, the barrier and pro-
tective effect of APHP/DOPO is weaker than that of APHP on ther-
mosets. The reason is that the expanded char layer of APHP/DOPO/
EP did not hinder heat spreading more effectively due to the thin
char film with hole. Whereas the compact and sealed char layer
from APHP is more effective on protecting the inner matrix during
combustion.
Additionally, the element contents of the residues after suffi-
ciently mixed and grinded from the cone calorimeter tests were
detected via Energy Dispersive Spectrometer (EDS). Then the
reserved and released phosphorus(P) contents were calculated and
shown in Table 5. In Table 5, the P contents of the 2%APHP/4%DOPO/
EP residues were reserved more in residue and released less in gas,
indicating that the high char yields were caused by the action of
phosphorus contents and the less phosphorus contents in gas
phase also worked in high efficiency. The results also disclosed why
Charring effect (%) Barrier and protective effect (%)
17.4 28.2
21.8 30.1
14.4 28.6
9.2 15.4
178 J. Wang et al. / Polymer Degradation and Stability 130 (2016) 173e181

Fig. 3. Digital photos of thermoset residue from cone calorimeter. (a) 2%APHP/4%DOPO/EP; (b) 6%APHP/EP; (c) 6%DOPO/EP; (d) EP.

Fig. 4. SEM photos of the residues from cone calorimeter. (a) 2%APHP/4%DOPO/EP; (b) 6%APHP/EP; (c) 6%DOPO/EP; (d) EP.
 J. Wang et al. / Polymer Degradation and Stability 130 (2016) 173e181
Table 5
Reserved and released P contents in cone calorimeter test by EDS.
Samples P Ratio in residues (%) Char yield (%)
EP 0 6.0
6%DOPO/EP 2.19 10.1
6%APHP/EP 4.66 15.0
2%APHP/4%DOPO/EP 4.86 16.8
6%DOPO/EP had better flame inhibition and 6%APHP/EP had better
charring and barrier and protective effects. But the more phos-
phorus contents of 2%APHP/4%DOPO/EP were reserved in residue
than those of 6%DOPO/EP and 6%APHP/EP. The results confirm that
phosphaphenanthrene group would react with APHP to form res-
idue, thereby enhancing the phosphorus contents in residue and
locking more carbon. That’s why the char yield rise. The more
phosphorus in residue verifies that DOPO and APHP will interact
and form the products with high phosphorus ratio to exert well
flame-retardant effect. The products contributed to inhibiting the
combustion process.
3.5. Thermal analysis
The synergistic flame retardant effects in oxidation condition
have been evaluated using cone calorimeter. The direct interaction
of APHP/DOPO in thermosets can be investigated using TGA in ni-
trogen without oxygen in Fig. 5. Therefore, it has not the influence
of oxidation reaction during the interaction of all the compositions.
In Fig. 5, DOPO caused the early decomposition of thermoset at the
lower temperature and produced more residue comparing with
pure thermoset. Differently, APHP sustained initial decomposition
temperature of thermoset at high temperature and promoted the
formation of more residue than DOPO did. When DOPO and APHP
were incorporated commonly into epoxy thermosets in a certain
ratio, DOPO/APHP/EP thermosets inherited the similar initial
decomposition temperature from DOPO. With the increased tem-
perature and decomposition of APHP, the DOPO/APHP/EP thermo-
sets showed the decreased decomposition speed comparing with
both DOPO/EP and APHP/EP. The phenomenon implies that the
interaction between DOPO and APHP during heating caused more
carbon content locked in residue. At last, the residue yield of 2%
APHP/4%DOPO/EP also kept the same level as that of 6%APHP/EP
Fig. 5. TGA curves of APHP, DOPO and the typical thermosets.
179
Initial P ratio in samples (%) Reserved P ratio in total P (%) Released P ratio in total P (%)
0 0 0
0.86 25.6 74.4
1.22 57.4 42.6
0.98 83.7 16.3
although only 2%APHP with higher charring ability was loaded. The
results reveal that the interaction between APHP and DOPO caused
the synergistic charring effect.
3.6. Pyrolysis-gas chromatography-mass spectrometry (Py-GC/MS)
of the flame-retardant EP thermosets
To further explore the pyrolysis behavior and its synergistic
flame-retardant mechanism of APHP/DOPO in EP thermosets, the
Py-GC-MS was adopted with a pyrolysis temperature at 550  C.
Fig. 6 shows GC chromatograms of the 6%DOPO/EP, 6%APHP/EP and
synergistic 2%APHP/4%DOPO/EP thermosets, and the determined
chemical structures corresponding to the GC peaks are listed in
Table 6. Comparing with the control sample 6%DOPO/EP and 6%
APHP/EP, the GC of 2%APHP/4%DOPO/EP thermoset shows
decreased intensity of peaks (a), (b), (c), and (d), and almost same
intensity of peaks (e), (f) and (g) appears in latter retention time.
The results reveal that APHP/DOPO commonly acted on the
decomposition of the matrix and caused the reduction of smaller
fragments. The (a), (b), (c), and (d) peaks corresponded to ethylene
oxide, phenol and its derivatives produced by decomposition of
bisphenol A part, whereas the larger fragments of peaks (e), (f), and
(g) corresponded to bisphenol A derivatives. In combination with
the decreasing increasing CO ratio and the reduction of av-EHC
value in cone calorimeter test, the joint action of APHP/DOPO led
to the reduction of smaller fragments release and the large frag-
ments were easily quenching and difficult to fully burning. Further,
the larger fragments can be quenched by the less PO contents
because of the less free radical spots, accordingly the flame inhi-
bition effect of APHP/DOPO still was increased although more
phosphorus contents were reserved in residue. In addition, the
Fig. 6. GC chromatograms of flame-retardant EP thermosets after pyrolysis.
180 J. Wang et al. / Polymer Degradation and Stability 130 (2016) 173e181
Table 6
Chemical structures of pyrolysis fragments in GC of flame-retardant thermosets determined by MS.
No. Chemical structure m/z
a 44
b 94
c 136
d 134
Fig. 7. Synergistic flame-retardant mechanism from APHP and DOPO.
more large fragments are main components of the char residue,
which also should enhance the charring effect in condensed phase.
3.7. Flame-retardant mechanism: synergistic effect between APHP
and DOPO
According to all the discussed results, the synergistic flame-
retardant mechanism of APHP/DOPO in EP thermosets can be
concluded in Fig. 7. After the thermosets was ignited, DOPO firstly
decomposed to generate phosphorus acid before APHP and catalyze
the pyrolysis of matrix, and then APHP began to decompose. The two
flame retardants interacted each other and also co-worked on ma-
trix to lock more carbon contents [14,27]. The phosphorus-rich
tougher residue blending with aluminum phosphate was formed.
The higher char yield and higher expanding ratio of char layer
reduced fuel and kept the barrier effect in a high level, which exerts
well synergistic effect between APHP and DOPO in condensed phase.
In gaseous phase, DOPO firstly released PO free radicals to exert
flame inhibition effect. APHP also released the phosphorus-
containing fragments to inhibit the flame. Since the less small py-
rolysis fragments were released and the large fragments were
easily quenched during combustion because of the less free radical
spots, DOPO/APHP system brought outstanding flame inhibition
effect. Hence, the better synergistic flame inhibition effect also
appeared in gaseous phase.
No. Chemical structure m/z
e 228
f 252
f 267
g 266
4. Conclusion
APHP/DOPO flame-retardant system was applied in epoxy resin.
Comparing with the thermosets added APHP and DOPO alone in a
same addition mass fraction, the LOI value and UL94 rating all rise,
and the HRR and THR values all reduced obviously. All these results
confirm the synergistic flame-retardant effect between APHP and
DOPO. During combustion, APHP/DOPO system jointly worked and
reduced the small flammable pyrolysis fragments. Accordingly, the
large pyrolysis fragments are easy to quenched, thereby obtaining
the better flame inhibition effect. APHP/DOPO system also jointly
worked to reserve more phosphorus-rich char, which testifies the
higher charring effect. The more phosphorus-rich residue induced
by APHP/DOPO form a more expanded char layer, indicating a
synergistic barrier and protective effect. In consequence, APHP/
DOPO system can endow EP thermosets with the better flame
retardancy by their synergistic effect in both gaseous phase and
condensed one.
Acknowledgments
Financial support was provided by key project of science and
technology plan from Beijing Municipal Education Commission
&Beijing Nature Science Foundations (No. KZ201510011009) and
Graduate Student Scientific Research Ability Promotion Plan (2015).
 J. Wang et al. / Polymer Degradation and Stability 130 (2016) 173e181 181

References [23] S. Yang, J. Wang, S.Q. 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) 63e69.
[1] B.W. Liu, H.B. Zhao, Y. Tan, L. Chen, Y.Z. Wang, Novel crosslinkable epoxy
[24] A. Toldy, B. Szolnoki, G.Y. Marosi, Flame retardancy of fibre-reinforced epoxy
resins containing phenylacetylene and azobenzene groups: from thermal
resin composites for aerospace applications, Polym. Degrad. Stab. 96 (3)
crosslinking to flame retardance, Polym. Degrad. Stab. 122 (2015) 66e76.
(2011) 371e376.
[2] G.M. Wu, B. Schartel, H. Bahr, M. Kleemeier, D. Yu, A. Hartwig, Experimental
[25] B. Perret, B. Schartel, K. Sto € ß, M. Ciesielski, J. Diederichs, M. Do
€ring, J. Kra€mer,
and quantitative assessment of flame retardancy by the shielding effect in
V. Altsta€dt, Novel DOPO-based flame retardants in high-performance carbon
layered silicate epoxy nanocomposites, Combust. Flame 159 (12) (2012)
fibre epoxy composites for aviation, Eur. Polym. J. 47 (5) (2011) 1081e1089.
3616e3623.
[26] X.Q. Liu, J.Y. Liu, S. Sun, J. Chen, S.J. Cai, Novel flame-retardant epoxy based on
[3] A. Dasari, Z.Z. Yu, G.P. Cai, Y.W. Mai, Recent developments in the fire retard-
zinc methylethyl phosphinate, Fire Mater. 39 (5) (2014) 599e608.
ancy of polymeric materials, Prog. Polym. Sci. 38 (9) (2013) 1357e1387.
[27] L.J. Qian, Y. Qiu, J.Y. Wang, W. Xi, High-performance flame retardancy by char-
[4] P. Mohan, A Critical Review: the modification, properties, and applications of
cage hindering and free radical quenching effects in epoxy thermosets,
epoxy resins, Polym. Plast. Technol. 52 (2) (2013) 107e125.
Polymer 68 (2015) 262e269.
[5] D. Pinto, L. Bernardo, A. Amaro, S. Lopes, Mechanical properties of epoxy
[28] X. Wang, Y. Hu, L. Song, H.Y. Yang, W.Y. Xing, H.D. Lu, Synthesis and char-
nanocomposites using titanium dioxide as reinforcement-A review, Constr.
acterization of a DOPO-substituted organophosphorus oligomer and its
Build. Mater. 95 (2015) 506e524.
application in flame retardant epoxy resins, Prog. Org. Coat. 71 (1) (2011)
[6] M.W. Szyndler, J.C. Timmons, Z.H. Yang, A.J. Lesser, T. Emrick, Multifunctional
72e82.
deoxybenzoin-based epoxies: synthesis, mechanical properties, and thermal
[29] M. Gao, S.S. Yang, J. Appl. Polym. Sci. 115 (4) (2010) 2346e2351.
evaluation, Polymer 55 (17) (2014) 4441e4446.
[30] S. Levchik, A. Piotrowski, E. Weil, Q. Yao, A novel intumescent flame-retardant
[7] H.C. Chang, T.L. Hung, C.H. Lin, W.C. Su, Facile preparation of a phosphinated
epoxy resins system, Polym. Degrad. Stab. 88 (1) (2005) 57e62.
bisphenol and its low water-absorption epoxy resins for halogen-free copper
[31] A. Scha €fer, S. Seibold, O. Walter, M. Do €ring, Novel high Tg flame retardancy
clad laminates, Polym. Degrad. Stab. 98 (1) (2013) 102e108.
approach for epoxy resins, Polym. Degrad. Stab. 93 (2) (2008) 557e560.
[8] L.Z. Liu, H. Zhang, N. Zhang, W. Ling, The preparation and application of CTBN
[32] L.J. Qian, Y.J. Ye, Y. Qiu, S.R. Qu, Thermal degradation behavior of the com-
modified epoxy adhesive, Pigm. Resin Technol. 44 (6) (2015) 358e363.
pound containing phosphaphenanthrene and phosphazene groups and its
[9] B. Sood, M. Pecht, Conductive filament formation in printed circuit boards:
flame retardant mechanism on epoxy resin, Polymer 52 (24) (2011)
effects of reflow conditions and flame retardants, J. Mater. Sci. Mater. El 22
5486e5493.
(10) (2011) 1602e1615.
[33] W.C. Zhang, X.M. Li, R.J. Yang, Study on flame retardancy of TGDDM epoxy
[10] S.Q. Ma, W.Q. Liu, N. Gao, Z.L. Yan, Y. Zhao, Synthesis and properties of LED-
resins loaded with DOPO-POSS compound and OPS/DOPO mixture, Polym.
packaging epoxy resin toughened by a novel polysiloxane from hydrolysis
Degrad. Stab. 99 (2014) 118e126.
and condensation, Macromol. Res. 19 (9) (2011) 972e979.
[34] B. Perret, B. Schartel, K. Sto € ß, M. Ciesielski, J. Diederichs, M. Do
€ring, J. Kra€mer,
[11] N. Gao, W.Q. Liu, S.Q. Ma, C.Y. Tang, Z.L. Yan, Cycloaliphatic epoxy resin
V. Altsta€dt, A New halogen-free flame retardant based on 9,10-dihydro-9-oxa-
modified by two kinds of oligo-fluorosiloxanes for potential application in
10-phosphaphenanthrene-10-oxide for epoxy resins and their carbon fiber
light-emitting diode (LED) encapsulation, J. Polym. Res. 19 (8) (2012) 9923.
composites for the automotive and aviation industries, Macromol. Mater. Eng.
[12] W. Rmili, M.P. Deffarges, F. Chalon, Z. Ma, R. Leroy, Dynamic mechanical
296 (1) (2011) 14e30.
properties and thermal effect of an epoxy resin composite, encapsulation’s
[35] Y. Qiu, L.J. Qian, W. Xi, X.X. Liu, Joint flame-retardant effect of triazine-rich and
element of a new electronic component, J. Electron. Mater. 43 (3) (2014)
triazine/phosphaphenanthrene compounds on epoxy resin thermoset, J. Appl.
702e707.
Polym. Sci. 133 (14) (2016) 43241.
[13] S.Q. Huo, J. Wang, S. Yang, B. Zhang, Y.S. Tang, A phosphorus-containing
[36] L. Zang, S. Wagner, M. Ciesielski, P. Müller, M. Do €ring, Novel star-shaped and
phenolic derivative and its application in benzoxazine resins: curing behavior,
hyperbranched phosphorus-containing flame retardants in epoxy resins,
thermal, and flammability properties, J. Appl. Polym. Sci. 133 (19) (2016)
Polym. Adv.Advan. Technol. 22 (7) (2011) 1182e1191.
43403.
[37] T. Kawahara, A. Yuuki, K. Hashimoto, K. Fujiki, T. Yamauchi, N. Tsubokawa,
[14] L.J. Qian, Y. Qiu, N. Sun, M.L. Xu, G.Z. Xu, F. Xin, Y.J. Chen, Pyrolysis route of a
Immobilization of flame-retardant onto silica nanoparticle surface and prop-
novel flame retardant constructed by phosphaphenanthrene and triazine-
erties of epoxy resin filled with the flame-retardant-immobilized silica (2),
trione groups and its flame-retardant effect on epoxy resin, Polym. Degrad.
React. Funct. Polym. 73 (3) (2013) 613e618.
Stab. 107 (2014) 98e105.
€ring, Bridged phosphorylated [38] L.J. Qian, L.J. Ye, X.L. Han, G.Z. Xu, Y. Meng, Construction and properties of
[15] C. Klinkowski, S. Wagner, M. Ciesielski, M. Do
compounds based on phosphaphenanthrene group, Prog. Chem. 22 (9) (2010)
diamines: synthesis, thermal stability and flame retarding properties in epoxy
1776e1783.
resins, Polym. Degrad. Stab. 106 (2014) 122e128.
[39] B. Schartel, B. Perret, B. Dittrich, M. Ciesielski, J. Kr€ amer, P. Müller, V. Altst€
adt,
[16] L.J. Qian, Y. Qiu, J. Liu, F. Xin, Y.J. Chen, The flame retardant group-synergistic-
L. Zang, M. Do €ring, Flame retardancy of polymers: the role of specific reactions
effect of a phosphaphenanthrene and triazine double-group compound in
in the condensed phase, Macromol. Mater. Eng. 301 (1) (2016) 9e35.
epoxy resin, J. Appl. Polym. Sci. 131 (3) (2014) 39709.
[40] B. Schartel, A.I. Balabanovich, U. Braun, U. Knoll, J. Artner, M. Ciesielski,
[17] Q.Q. Luo, Y.C. Yuan, C.L. Dong, S.M. Liu, J.Q. Zhao, High performance fire-
M. Do €ring, R. Perez, J.K.W. Sandler, V. Altsta €dt, T. Hoffmann, D. Pospiech,
retarded epoxy imparted by a novel phenophosphazine-containing anti-
Pyrolysis of epoxy resins and fire behavior of epoxy resin composites flame-
flaming compound atultra-low loading, Mater. Lett. 169 (2016) 103e106.
retarded with 9,10-dihydro-9-oxa-10-phosphaphenanthrene-10-oxide addi-
[18] L.J. Qian, L.J. Ye, G.Z. Xu, J. Liu, J.Q. Guo, The non-halogen flame retardant
tives, J. Appl. Polym. Sci. 104 (4) (2007) 2260e2269.
epoxy resin based on a novel compound with phosphaphenanthrene and
[41] Z.X. Weng, T. Senthil, D.X. Zhuo, L.J. Song, L.X. Wu, Flame retardancy and
cyclotriphosphazene double functional groups, Polym. Degrad. Stab. 96 (6)
thermal properties of organoclay and phosphorous compound synergistically
(2011) 1118e1124.
modified epoxy resin, J. Appl. Polym. Sci. 133 (18) (2016) 43367.
[19] K. Shree Meenakshi, E. Pradeep Jaya Sudhan, S. Ananda Kumar,
[42] L. Zhang, Y.C. Wang, Q. Liu, X.F. Cai, Synergistic effects between silicon-
M.J. Umapathy, Development and characterization of novel DOPO based
containing flame retardant and DOPO on flame retardancy of epoxy resins,
phosphorus tetraglycidyl epoxy nanocomposites for aerospace applications,
J. Therm. Anal. Calorim. 123 (2) (2016) 1343e1350.
Prog. Org. Coat. 72 (3) (2011) 402e409.
[43] S. Liu, Z.P. Fang, H.Q. Yan, H. Wang, Superior flame retardancy of epoxy resin
[20] C.S. Wang, C.H. Lin, Novel phosphorus-containing epoxy resins. Part II: curing
by the combined addition of graphene nanosheets and DOPO, Rsc. Adv. 6 (7)
kinetics, Polymer 41 (24) (2000) 8579e8586.
(2016) 5288e5295.
[21] M. Raphaёl, N. Claire, F. Laurent, S. Rodolphe, D. Ghislain, Synthesis of bio-
[44] J.Y. Wang, L.J. Qian, B. Xu, W. Xi, X.X. Liu, Synthesis and characterization of
based phosphorus-containing flame retardants for epoxy thermosets com-
aluminum poly-hexamethylenephosphinate and its flame-retardant applica-
parison of additive and reactive approaches, Polym. Degrad. Stab. 120 (2015)
tion in epoxy resin, Polym. Degrad. Stab. 122 (2015) 8e17.
300e312.
[45] S. Brehme, B. Schartel, J. Goebbels, O. Fischer, D. Pospiech, Y. Bykov, M. Do €ring,
[22] S. Yang, J. Wang, S.Q. Huo, L.F. Cheng, M. Wang, Preparation and flame
Phosphorus polyester versus aluminium phosphinate in poly(butylene tere-
retardancy of an intumescent flame-retardant epoxy resin system constructed
phthalate) (PBT): flame retardancy performance and mechanisms, Polym.
by multiple flame-retardant compositions containing phosphorus and nitro-
Degrad. Stab. 96 (5) (2011) 875e884.
gen heterocycle, Polym. Degrad. Stab. 119 (2015) 251e259.