Microencapsulation of ammonium polyphosphate with melamine-formaldehyde-tris(2-hydroxyethyl)isocyanurate resin and its flame retardancy in polypropylene

Article abstract of DOI:10.1039/c5ra14586d

Microencapsulated ammonium polyphosphate with a novel melamine-formaldehyde-tris(2-hydroxyethyl)isocyanurate resin (MFT resin) was prepared by in situ polymerization. This microencapsulated ammonium polyphosphate was named MFT-APP. The MFT resin and MFT-APP were characterized by FTIR, XPS, TGA and SEM. Polypropylene (PP)/MFT-APP and PP/MFT-APP/THEIC were prepared. Their thermal stability and flame retardancy were investigated through TGA, limiting oxygen index (LOI), vertical burning tests and cone calorimeter tests. SEM was used again to observe the morphology of char residues. The results illustrated that MFT-APP and MFT-APP/THEIC had excellent flame retardant properties on PP. Furthermore, a water resistant test demonstrated that PP/MFT-APP/THEIC had better flame durability than PP/APP/THEIC. TGA-FTIR and XPS results revealed that the efficient flame retardancy of MFT-APP/THEIC was mainly attributed to the formation of a compact and thermostable intumescent char on underlying materials during a fire.

Full text of DOI:10.1039/c5ra14586d

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Microencapsulation of ammonium polyphosphate
with melamine-formaldehyde-tris(2-hydroxyethyl)
isocyanurate resin and its flame retardancy in
polypropylene
Cite this: RSC Adv., 2015, 5, 88445
*
Ziwei Jiang and Gousheng Liu
Microencapsulated ammonium polyphosphate with a novel melamine-formaldehyde-tris(2-hydroxyethyl)
isocyanurate resin (MFT resin) was prepared by in situ polymerization. This microencapsulated
ammonium polyphosphate was named MFT–APP. The MFT resin and MFT–APP were characterized by
FTIR, XPS, TGA and SEM. Polypropylene (PP)/MFT–APP and PP/MFT–APP/THEIC were prepared. Their
thermal stability and flame retardancy were investigated through TGA, limiting oxygen index (LOI),
vertical burning tests and cone calorimeter tests. SEM was used again to observe the morphology of
char residues. The results illustrated that MFT–APP and MFT–APP/THEIC had excellent flame retardant
properties on PP. Furthermore, a water resistant test demonstrated that PP/MFT–APP/THEIC had better
flame durability than PP/APP/THEIC. TGA-FTIR and XPS results revealed that the efficient flame
retardancy of MFT–APP/THEIC was mainly attributed to the formation of a compact and thermostable
intumescent char on underlying materials during a fire.
Received 23rd July 2015
Accepted 12th October 2015
DOI: 10.1039/c5ra14586d
www.rsc.org/advances
nd that APP is easily exuded to the surface of materials,
resulting in
1. Introduction
a
decrease of ame-retardant efficiency.13,14
Furthermore, APP has a low compatibility with polymer mate-
rials, affecting mechanical properties of polymeric compos-
ites.15,16 Modication of APP with surfactants or
microencapsulating with water-insoluble polymers are effective
methods to improve water resistance and the polarity of APP.17,18
Melamine-formaldehyde (MF) resin is widely used as coating
materials for modication. For example, Wu19 studied the ame
retardancy of microencapsulated ammonium polyphosphate
with MF resin on PP, he found that the particles' size and water
absorption of MFAPP was decreased. Yang20 studied the thermal
stability and the ame retardancy of PP/MF-APP/pentaerythritol
(PER) system, he found that MF-APP was benecial for
improving the thermal stability and ame retardancy of PP/MF-
APP/PER composites. However, researchers also found that APP
or MF-APP used alone in PP cannot reach UL-94 V-0 rating at
30% dosage.19,21 Incorporating an effective charring agent into
the coating materials will be a promising route.
Tris(2-hydroxyethyl)isocyanurate (THEIC) was mainly used
to prepare the polyester-based heat-resisting insulating paint.
THEIC also could be a modier for MF resin which could
improve the exibility and cohesiveness of MF resin.22–24 As
a derivative of triazine compound, THEIC and its derivatives
could be served as the charring agent. Bourbigot25 found that
there was a synergistic effect between APP and THEIC. Chen26
studied the advantages and disadvantages of THEIC when it was
served as a charring agent. Yuan27 and Chen28 synthesized
With the development of materials science, polymer materials
have received high speed development. As one of the widely
used polymer materials, polypropylene (PP) has outstanding
features such as low price, low density, non-toxicity, easy pro-
cessing and excellent mechanical properties, so it can be used
in electrical, building, transport applications and as general
household materials.1,2 However, its inherent high ammability
and melt dripping drawbacks limit its application range.
Therefore, it is essential to endow PP with good ame
retardancy.3
In recent years, intumescent ame retardants (IFRs) have
drawn much attention due to their halogen-free, low toxicity,
low smoke production and anti-dropping characteristics.4,5
Generally, a typical IFR system is composed of three compo-
nents: an acid resource, a charring agent and a blowing agent.6–8
The acid resource is generally inorganic acid or precursor of
acid, for example, ammonium polyphosphate (APP).9 During
the combustion process, APP can release the NH3 and the
inorganic acid, and then there is a esterication reaction
between the inorganic acid and the charring agent. At last it will
form a intumescent charred layer cover on the surface of the
polymer materials.10–12 During materials processing, workers
State Key Laboratory of Chemical Engineering, East China University of Science and
Technology, Shanghai 200237, China. E-mail: gsliu@ecust.edu.cn; Tel:
+86-21-33612485
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 88445–88455 | 88445

View Article Online
RSC Advances
Paper
a novel charring agent tris(2-hydrooxyethyl)isocyanurate ter- added into a four-neck bottle with a stirrer. The mixture was
ephthalic acid ester, this charring agent exhibited high ame adjusted to pH 8–9 with 10% Na2CO3 solution, heated to about
retardant efficiency on polylactide (PLA) and PP. Theoretically, 95 ꢀC and kept at that temperature for 0.5 h. When the mixture
introducing THEIC into MF resin as the coating materials for turned into transparent, the suitable amount of THEIC was put
APP will be a novel route for improving the water resistance, into the mixture and kept at 95 ꢀC for another 1 h. The MFT pre-
thermal stability and ame retardancy of APP. In this work, MFT polymer solution was prepared. Fig. 1 shows reaction scheme.
resin was prepared by in situ polymerization and it was char-
Preparation of microencapsulated APP. APP (100 g) was rst
acterized by Fourier transform infrared (FTIR). Then MFT resin dispersed in 200 ml ethanol. Then suitable amount of MFT pre-
was used as the coating material for APP. The structure and polymer solution (coating rate is 30%) was added into the
thermal property of MFT–APP were characterized by FTIR, X-ray suspension, the pH of the mixture was adjusted to 4–5 with
ꢀ
photoelectron spectroscopy (XPS), thermogravimetric analysis acetic acid. The resulting mixture was heated at 80 C for 2 h.
(TGA) and scanning electron microscope (SEM). The MFT–APP Then, the mixture was cooled to room temperature, ltered,
and MFT–APP/THEIC were added into PP, measurements such washed with distilled water, and dried at 100 ꢀC. The MFT–APP
as TGA, LOI, vertical burning test (UL-94) and cone calorimeter powder was nally obtained.
test were used. The morphology of char residues were observed
by SEM. Water resistance of the ame retardant composites was
2.3. Preparation of ame-retardant PP samples
investigated. TGA-FTIR and XPS were used to analysis the
ꢀ
All samples were prepared via melt compounding at 180 C in
an internal mixer (SU70, Suyan Technology Company, China)
with a rotate speed of 20 rpm for 8–10 minutes for each sample.
The resulting composites were dried in an oven, and then
molded into testing bars with plate vulcanization machine for
re property characterization.
thermal degradation of ame retardant composites.
2. Experimental section
2.1. Materials
Polypropylene (PP, isotactic, T30 s, melt-ow rate: 2.3 g/10 min)
was supplied by SINOPEC Maoming Company, China. Ammo-
nium polyphosphate (APP) with average particle size 10 mm was
purchased from Shandong Shi'an Chemical Co., Ltd. THEIC
was purchased from Changzhou Lantian Chemical Co., Ltd.
Melamine was obtained from Sinopharm Chemical Reagent
Co., Ltd. Formaldehyde solution (mass fraction 36%) was ob-
tained from Meryer (Shanghai) Chemical Technology Co., Ltd.
Anhydrous sodium carbonate (Na2CO3) was obtained from
Tianjin Zhiyuan Chemical Reagents, China. Acetic acid was
purchased from Sinopharm Chemical Reagent Co., Ltd. All
reagents were used as received.
2.4. Characterization
FTIR spectra of the samples were recorded on a Bruker-Vertex
70 FTIR spectrometer in the range of 400–4000 cmꢁ1 by using
KBr pellets. The thermogravimetry analysis (TGA) tests were
carried out onꢁD1iamond TG/DTG (PE, America) at a heating rate
of 10 ꢀC min from room temperature to 800 ꢀC with air
atmosphere, sample weight was kept within 10 mg. The X-ray
photoelectron spectroscopy (XPS) spectra were recorded with
a VG ESCALAB MK II spectrometer using Al Ka excitation
radiation (hn ¼ 1253.6 eV). Water contact angles were measured
with a JC2000C2 contact angle goniometer (Shanghai Zhong-
chen Powereach Company, China) by the sessile drop method.
2.2. Preparation of microencapsulated APP
Preparation of the MFT resin pre-polymer. 37% formalde- Limiting oxygen index (LOI) was carried out in an HC-2 oxygen
hyde solution and melamine (with a mole ratio of 3 : 1) were index meter (Jiangning Analysis Instrument Co., China). The
Fig. 1 The reaction scheme of MFT resin pre-polymer.
88446 | RSC Adv., 2015, 5, 88445–88455
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
vertical burning tests were conducted according to the UL-94 characteristic peaks in THEIC indicated the successful incor-
test standard (ASTM D3801) with a test dimensions of 130 ꢂ poration of THEIC into the MF resin.
13 ꢂ 3 mm3. The morphology of the residual char obtained
3.1.2. Thermal stability of MF and MFT resin. The TG and
from the LOI test was examined by means of eld emission DTG curves of MF and MFT resin are shown in Fig. 3. MF and
scanning electron microscopy (JSM-6360). The surface of MFT resin show a similar decomposing behavior. There were
residual char was sputter-coated with a gold layer before two main decomposing processes for them in air atmosphere:
examination. The combustion properties were evaluated by 50–100 ꢀC and 350–400 ꢀC. We supposed the rst weight loss is
a cone calorimeter. All samples (100 ꢂ 100 ꢂ 3 mm3) were primarily due to water vaporization. The second weight loss is
exposed to a Stanton Redcro cone calorimeter under a heat correspond to the structural decomposition of the resins. The
ux of 50 kW mꢁ2 according to ISO-5660 standard. Tensile tests maximum weight loss rate of MF resin is larger than MFT resin.
were completed in accordance with the procedures in GB/T The DSC results for MF and MFT resin are shown in Fig. 4. For
1040-1992 at a crosshead speed of 50 mm minꢁ1. Flexural them, there is a main endothermic peak at 90 ꢀC, which is
properties were carried out in accordance with the procedures correlated with the water vaporization. With the rising of
in GB/T 9314-2000 at a crosshead speed of 2 mm minꢁ1
.
temperature, endothermic situation is declining. Though there
Thermogravimetry-Fourier transform infrared spectroscopy is a some small up and downs, we think the lowest point (about
ꢀ
(TGA-FTIR) was performed using a SDT Q600 V20.5 Build 15 150 C) may be the curing temperature for MFT resin.
instrument that was interfaced to a Nicolet IS10 FTIR spec-
3.1.3. Chemical structure of MFT–APP. The FTIR spectra of
trometer. About 5.0 mg of the sample was put in an alumina APP and MFT–APP were shown in Fig. 5, the spectrum of MFT
crucible and heated from 40 to 800 ꢀC. The heating rate was set resin was added as a contrast. The typical absorption peaks of
as 20 C minꢁ1 (air atmosphere).
APP included 3223 cmꢁ1 (N–H), 1256 (P]O), 1069 cmꢁ1 (P–O
asymmetric vibration), 1020 cmꢁ1 (symmetric vibration of PO2),
881 cmꢁ1 (P–O asymmetric stretching vibration), and 800 cmꢁ1
(P–O–P). For the MFT–APP, the main absorption peaks
appeared at 3204, 1554, 1419, 1255, 1069, 1020, 880 and 800
cmꢁ1. The typical absorption peaks of APP were well veried in
the structure of MFT–APP. Meanwhile, the peak at 1554 cmꢁ1
and 1419 cmꢁ1 were assigned to the ring vibration of MEL from
the MF resin. It was clear that not only the absorption peaks of
MFT resin but also the characteristic bands of APP appeared on
the FTIR spectra of MFT–APP. Above results proved that APP
was well coated by the MFT resin.
The XPS spectrum of APP and MFT–APP was shown in Fig. 6.
The peaks located at 134 eV and 192 eV were assigned to the
typical chemical shis of P2p and P2s in APP. For MFT–APP, the
intensities of peaks mentioned above decreased sharply,
meanwhile the intensities of the C1s peak increases greatly. The
intensities of the N1s peak did not increase greatly like the C1s
ones, which were attributed to the structure of THEIC in the
MFT resin. The growth and decline relationship among these
typical chemical shis eloquently proved that APP was well
coated by the MFT resin.
ꢀ
3. Results and discussion
3.1. Characterization of MFT–APP
3.1.1. Chemical structure of MFT resin. FTIR spectra of MF
resin and MFT resin were shown in Fig. 2. For MF resin, the
absorption peak at 3332 cmꢁ1 was attributed to the stretching
vibration of N–H and O–H, or coupled absorption of hydrogen
bonds. The peak at 2943 cmꢁ1 was attributed to the C–H
stretching vibration of –CH2 groups. The absorptions of 1589,
1500, and 1360 cmꢁ1 were due to the ring vibration of MEL. The
spectrum of MFT resin revealed not only well-dened absorp-
tion peaks of MF resin, but also the characteristic bands of
THEIC. There were two obvious absorption peaks at 1713 cmꢁ1
and 766 cmꢁ1, which could be found in the spectrum of MFT
resin. They were assigned to the –C]O stretching vibration and
triazine ring skeleton vibration. The appearance of these
Fig. 7 shows the surface morphologies of APP and MFT–APP.
The surface of APP particle was very smooth. Aer APP was
microencapsulated by MFT resin, the MFT–APP presented
a comparably rough surface. It was evidence that APP was well
coated by the MFT resin. As we know, water contact angle tests
could evaluate the surface hydrophobicity of solid powder. As
shown in Fig. 7, the water contact angle of APP is only 9ꢀ. This
indicated that pure APP is hydrophobic with high surface
energy and easily attacked by the inuence of water or moisture.
However, due to the formation of the weak polar MFT resin
microcapsules on the surface of APP, the hydrophile of MFTAPP
is greatly decreased. In this condition, the water contact angle of
MAPP reaches 45ꢀ. When MFT–APP particles are incorporated
into polymer matrix, the compatibility between MFT–APP and
polymer materials will be improved.
Fig. 2 FTIR spectrum of MF resin and MFT resin.
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 88445–88455 | 88447

View Article Online
RSC Advances
Paper
Fig. 3 TG and DTG curves of MF and MFT resin.
Table 1 shows the solubility of APP and MFT–APP. The
solubility of APP was 0.55 g/100 ml H2O at room temperature
(25 ꢀC). Aer APP particles were microencapsulated by the MFT
resin, the solubility of MFT–APP was decreased to 0.3 g/100 ml
H2O. The solubility was decreased about 45.5% aer APP was
microencapsulated. The results proved that APP was well
microencapsulated by MFT resin.
The TG and DTG curves of APP, MFAPP and MFT–APP were
shown in Fig. 8 and the corresponding data were listed in
Table 2. There were two decomposing processes for APP in air
atmosphere: 220–460 ꢀC and 460–800 ꢀC. The temperatures
with the maximum mass loss rates during the two steps are
330 ꢀC and 607 ꢀC, respectively. The weight loss of the rst stage
(220–460 ꢀC) should be attributed to the elimination of NH3 and
H2O in the thermal degradation process of polyphosphate. The
ꢀ
weight loss of the second stage (beyond 460 C) was attributed
Fig. 4 DSC of MF and MFT resin.
to the release of phosphoric acid, polyphosphoric acid, and
metaphosphoric acid with APP decomposition. However,
Fig. 5 FTIR spectrum of APP and MFT–APP.
88448 | RSC Adv., 2015, 5, 88445–88455
Fig. 6 XPS spectrum of APP and MFT–APP.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
Fig. 7 SEM of APP and MFT–APP. (a1) APP ꢂ 200, (a2) APP ꢂ 500, (a3) APP ꢂ 3000. (b1) MFT–APP ꢂ 200, (b2) MFT–APP ꢂ 500, (b3) MFT–APP ꢂ
3000.
MFAPP and MFT–APP showed different degradation behavior compounds would inhibit APP further decomposition. From
compared with APP at the whole process. There were about data listed in Table 2, it was clear that the thermal stability of
three degradation stages for MFAPP and MFT–APP. The onset MFT–APP was better than others in the high temperature. As
decomposition temperature (T5%) and the rst maximum mentioned above, THEIC can reacted with APP, introducing
weight loss temperature (Tmax1) were shied to lower tempera- THEIC into the MF resin would be helpful for improving the
ture. At the rst degradation stage, MFAPP and MFT–APP thermal stability of APP. Moreover, MFT–APP had about 34.9%
decomposed faster than APP, which can be attributed to the low residue aer 800 ꢀC, whereas the residue for APP and MFAPP at
thermal stability of the MF resin and MFT resin. The weight loss this temperature was just 14.5% and 24.2%, respectively. It is
might be caused by the release of the nonammable gases from clear that MFT–APP presented better thermal stability than APP
the resin. For the second and third degradation stages, it was and MFAPP in the high temperature.
the results of joint action between APP and the resin. The
residual of the resin and the phosphorus-containing
3.2. Flame retardancy and burning behaviors
3.2.1. LOI and UL-94 tests. The LOI and UL-94 tests were
listed in the Table 3.
Table 1 Solubility of APP and MFT–APP
It could be found that pristine PP (PP0) was highly
combustible and could not reach any UL-94 rating. Aer 30 wt%
APP and MFAPP was added into the PP (PP1 and PP2), the LOI
value increased to 21% and 30.5% respectively. However, PP1
and PP2 still could not pass the UL-94 V-0 test owing to the
Sample
Solubility (100 g mlꢁ1 H2O)
APP
MFT–APP
0.55
0.30
Fig. 8 The TG and DTG curves of APP, MFAPP and MFT–APP.
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 88445–88455 | 88449

View Article Online
RSC Advances
Paper
Table 2 The data of TG and DTG test of APP, MFAPP and MFT–APP
Residual
600 ꢀC/%
Sample
T5%/ꢀC
Tmax1/ꢀC
Tmax2/ꢀC
Tmax3/ꢀC
Umax%/min
700 ꢀC/%
800 ꢀC/%
APP
MFAPP
MFT–APP
320.4
199.8
199.8
330
310
308
—
405
391
607
549
524
ꢁ8.13
ꢁ3.23
ꢁ3.19
49.2
36.9
42.2
23
29.2
37.6
14.5
24.2
34.9
Table 3 The LOI and UL-94 test results of the samplesa
Sample
PP (%)
APP (%)
MFAPP (%)
MFT–APP (%)
THEIC (%)
MFT–APP:THEIC
LOI (%)
UL-94
PP0
PP1
PP2
PP3
PP4
PP5
PP6
PP7
100
70
70
70
70
70
70
70
70
70
75
75
75
75
75
0
30
0
0
0
0
0
0
0
22.5
0
0
0
0
0
0
30
0
0
0
0
0
0
0
0
0
0
30
15
20
22.5
25
27
0
0
0
0
0
0
0
1
2
3
5
9
0
1
2
3
5
9
17.8
21
30.5
32
—
35
36
35.5
34
34.3
—
32.5
33
N.R
N.R
N.R
V-0
—
V-0
V-0
V-0
V-0
V-0
—
0
15
10
7.5
5
PP8
PP9
3
0
7.5
12.5
8.3
6.25
4.17
2.5
PP10
PP11
PP12
PP13
PP14
0
0
0
0
12.5
16.7
18.75
20.83
22.5
V-0
V-0
V-0
V-0
32
31
0
0
a
N.R ¼ no rating.
scarcity of charring agent. When 30 wt% MFT–APP was added to decomposition temperature (T5%) and similar decomposing
PP (PP3), PP3 successfully passed V-0 level of UL-94 test. The stages. The thermal stability of PP3 was better than PP2. When
reason may contain two aspects, one was the MFT resin released THEIC was incorporated into the PP/MFT–APP, the thermal
water vapor and NH3 gases which would reduce the concen- decomposition behavior of the ame retardant composites
tration of air and make the char swell to form protective char. changed greatly. Before temperature 630 ꢀC, PP6 had a lower
The other reason can be attributed to the structure of THEIC in thermal stability than PP2 and PP3, which could be attributed to
the MFT resin, which would enhance the synergistic effect the low stability of THEIC. At 287 ꢀC, the rst maximum weight
between the phosphorus and nitrogen. Though the MFT–APP/ loss was occurred for PP6. The rst maximum weight loss rate of
PP could pass the UL-94 V-0 test, the LOI value still be low. In PP6 was larger than PP2 and PP3. The reason might be THEIC
order to further improve the ame retardancy, THEIC was could promote the decomposition of MFT–APP. Aer tempera-
added into the PP/MFT–APP composites, which was served as ture 630 ꢀC, the thermal stability of PP6 was better than PP2 and
a charring agent. From the data listed in Table 3, it was clear PP3. The reason could be attributed to the synergistic effect
that the LOI values of the ternary composites were increased between THEIC and MFT–APP. The synergistic effect between
greatly. At the same additive level (30%), the highest LOI value THEIC and MFT–APP was benecial for forming a swell char
could obtain 36%. When the total addition of the MFT–APP and layer, and this swell char layer would protect the below mate-
THEIC was decreased into 25%, all ternary composites still rials from further decomposition.
could obtain UL-94 V-0 rating. These results demonstrated that
The residual charred layer was important for the ame
MFT–APP and MFT–APP/THEIC would remarkably improve retardancy of polymer materials. The images of residual chars
ame retardant performance of PP, and THEIC showed a good for PP1, PP2, PP3 and PP6 were shown in Fig. 10. As seen in
synergistic effect with MFT–APP.
Fig. 10a, no residual chars could be seen on the surface of PP1.
3.2.2. Thermal stability of ame retardant PP. The thermal It was attributed to lack of char formation during combustion.
degradation behavior of PP0, PP2, PP3 and PP6 was shown in Heat and volatiles could easily penetrate the surface during the
Fig. 9 and the corresponding data were listed in Table 4. It could burning process. For PP2 and PP3, a typical intumescent char-
be seen that only one step weight loss (300–450 ꢀC) for pure PP ring layer still could not be seen, which was due to insufficient
(PP0) and the maximum mass loss rate occurred at about charring agent introduced into the system. Though PP3 could
ꢀ
322.5 C. There was a similar thermal decomposition behavior pass the UL-94 V-0 test, the incorporation of THEIC into the MF
between PP2 and PP3, they had an approximate onset resin was helpful for improving the thermal stability of the
88450 | RSC Adv., 2015, 5, 88445–88455
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
Fig. 9 The TG and DTG curves of PP0, PP2, PP3 and PP6.
Table 4 The data of TG and DTG test of PP0, PP2, PP3 and PP6
Residual
600 ꢀC/%
Sample
T5%/ꢀC
Tmax1/ꢀC
Tmax2/ꢀC
Tmax3/ꢀC
Umax%/min
700 ꢀC/%
800 ꢀC/%
PP0
PP2
PP3
PP6
257.6
261
258.5
250.9
322.5
276
266
—
—
ꢁ16.95
ꢁ9.75
ꢁ7.77
ꢁ7.5
0
0
0
370
357
339
611
607
569
16.3
19.8
16.7
2.69
6.72
8.44
0.88
5.42
6.5
287
MFT–APP. The situation was improved greatly when THEIC dripping of PP. However, situation was improved to some extent
served as a charring agent. For PP6, a compact and swelling aer 30 wt% MFT–APP was added into the PP. For PP3, some
structure could be observed. Furthermore, there were lots of swollen residual char could be seen, though the degree of the
folds on the surface, which could act as a skeleton to strengthen expansion was not ideal. Aer THEIC was added into the ame
the surface layer.29,30
retardant composites, a compact and dense char layer was
This result provided further evidence that THEIC had a good observed on the external surface of PP6. This swollen charred
synergistic effect with the MFT–APP, homogeneous and stable layer could be effectively prevented the inside pyrolysis products
intumescent char layer formed, leading to better ame from transmitting into the ame zone.
retardancy.
The cone calorimeter test was considered to be an effective
The digital images of the residual chars for PP0, PP3 and PP6 way to simulate the actual re burning behavior and the
were shown in Fig. 11. Nothing was le for PP0 which was measurements provided several useful burning parameters.31–33
coincided with inherent properties of ammability and melt
Fig. 10 SEM image of the char from PP1, PP2, PP3 and PP6. (a1) PP1 ꢂ 500, (a2) PP1 ꢂ 1000. (b1) PP2 ꢂ 500, (b2) PP2 ꢂ 1000. (c1) PP3 ꢂ 500,
(c2) PP3 ꢂ 1000. (d1) PP6 ꢂ 500, (d2) PP6 ꢂ 1000.
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 88445–88455 | 88451

View Article Online
RSC Advances
Paper
the intumescent system based on MFT–APP and THEIC could
effectively improve the ame retardancy of PP.
In order to investigate water resistance of the ame retardant
composites, PP/APP/THEIC and PP/MFT–APP/THEIC were
soaked in water at 70 ꢀC for different time and their ame
retardancy properties were tested, including LOI and UL-94
tests. The results were listed in Table 6. PP/APP/THEIC only
obtained a V-2 rating aer 24 h, with melt dripping occurred
during combustion process. Aer 48 h, no rating could be ob-
tained for it. However, PP/MFT–APP/THEIC could still obtain
V-0 rating aer 96 h soaking time. The results indicated MFT–
APP/THEIC has a good water resistant property, the ame
retardant durability of PP/MFT–APP/THEIC was improved
greatly compared with PP/APP/THEIC composite.
The mechanical properties of ame retardant composites are
presented in Fig. 13, including tensile properties and exural
properties. It can be seen that the addition of APP into PP
decreases sharply the tensile strength and exural strength of
PP1, with a falling range about 35.1% and 21.4% respectively.
When MFT–APP is added into PP, the falling range is not so
much. The falling range for PP2 is only 18.9% and 0% respec-
tively. So there is a improvement of compatibility when MFT–
APP is added into PP compared with APP. When THEIC is added
into the composites, the situation between PP6 and PP9 is not
so obvious. The falling ranges of tensile strength and exural
strength of PP9 are 43.2% and 14.3% respectively. The results
for PP6 is 29.7% and 14.2%. We think the reason for the above
situation between PP6 and PP9 is that THEIC can react with APP
or MFT–APP in the processing. So the superiority of MFT–APP in
compatibility can not be shown.
Fig. 11 Digital photographs of residues for PP0, PP3 and PP6 after
cone calorimeter test.
The HRR and THR curves of PP0, PP3 and PP6 were shown in
Fig. 12 and the interrelated data were listed in Table 5. The HRR
of PP0 showed a signicant increase aer ignition up to the
peak value (pHRR) and decreased sharply toward the end of
burning. When 30 wt% MFT–APP or MFT–APP/THEIC were
added into PP, the HRR of the mixture decreased sharply. The
pHRR of PP3 and PP6 decreased to 375 kW mꢁ2 and 232
kW mꢁ2, respectively. The values were reduced by 69.7% and
81.2% compared to that of pure PP, respectively. The times to
pHRR (tpHRR) of PP3 and PP6 were 169 s and 192 s, respectively.
They were signicantly delayed in comparison with PP0. For
PP0, the largest THR value was 123.7 MJ mꢁ2. In contrast, the
curves of PP3 and PP6 showed a much lower value in THR plot,
about 116.4 MJ mꢁ2 and 100.7 MJ mꢁ2, respectively. For pSPR
and TSP, the relative relationship among PP0, PP3 and PP6 was
similar to that of pHRR and THR. At the end of burning, the
residual weight of PP3 and PP6 still had 17.4% and 20.7%,
respectively. The above results demonstrated that MFT–APP and
Fig. 12 Cone calorimeter test curves of PP0, PP3 and PP6 (a) heat release rate curves (HRR) (b) total heat release curves (THR).
Table 5 Combustion parameters obtained from cone calorimeter
Sample
TTI (s)
pHRR (kW mꢁ2
)
tpHRR (s)
THR (MJ mꢁ2
)
pSPR (m2 sꢁ1
)
TSP (m2)
Residual (%)
PP0
PP3
PP6
33
24
28
1238
375
232
149
169
192
123.7
116.4
100.7
0.154
0.058
0.04
15.78
14.45
13.38
0
17.4
20.7
88452 | RSC Adv., 2015, 5, 88445–88455
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
Fig. 14a, the small band observed at 3400–4000 cmꢁ1 can be
attributed to the O–H vibration of H2O; the intense bands at
2800–3100 cmꢁ1 are assigned to saturated hydrocarbons and
saturated alkane; another intense bands at 2300–2350 cmꢁ1 are
assigned to CO2; the bands at 2250–2300 cmꢁ1 are assigned to
CO which is transformed into CO2 with the rising of tempera-
ture; the bands at 1600–1900 cmꢁ1 are due to the structures
containing aromatic rings; the peak near 1370 cmꢁ1 is assigned
to gas containing the P–O–C structure; the sharp peaks at 930–
960 cmꢁ1 referred to the N–H vibration of NH3. In Fig. 14b, FTIR
spectra of volatile pyrolysis products evolved at different
temperature are shown. There is almost no infrared signal
Table
composites
6 Effect of water soaking time on LOI value of PP/IFR
PP/APP/THEIC
UL-94 V-rank
PP/MFT–APP/THEIC
Water treated
time
LOI
UL-94 V-rank
LOI
24 h
48 h
72 h
96 h
V-2
NR
NR
NR
30.2%
28.5%
27.5%
26.9%
V-0
V-0
V-0
V-0
33%
32%
32%
31%
3.3. Thermal degradation analysis
ꢀ
below 200 C, indicating that the MFT–APP/THEIC mixture do
TGA-FTIR spectra were used to analyze the gas produced during
the thermal degradation process of polymers. The three-
dimensional TGA-FTIR spectrum and characteristic spectra
obtained at different temperature of PP6 is shown in Fig. 14. In
not decompose under this temperature, thus it is thermostable
for melting-blend with PP (170–210 ꢀC). With further increasing
the temperature, the release of H2O, CO2, and NH3, etc. can be
detected. The reason for these gases production is that MFT–
APP can release NH3 to form poly(phosphorus acid) and these
acid can react with THEIC to release H2O. When the tempera-
ture increased to 400 ꢀC, a maximum signal intensity is
observed. These can be explained by that, with increasing
temperature, a swollen char is formed and prevented the poly-
mer from decomposing. Aerward, the signal intensity of the
pyrolysis products decreased gradually, indicating that the
decomposition rate of the composite is slowed down by the
promoted char formation. When the temperature above 500 ꢀC,
the intensity of CO2 is increasing gradually. The reason may be
the promoted char layer is decomposing.
In order to further analyze chemical structure of the char
layer, XPS spectra of C1s, O1s for PP6 aer combustion is
studied. The XPS spectra are shown in Fig. 15. The C1s spec-
trum is presented in the Fig. 15a. The peaks at 284.6 eV and
285.8 eV are assigned to C–H and C–C in aliphatic species and
C–O in ether, hydroxyl groups, C–O–P in hydrocarbonated
phosphate, and/or C–N in heterocyclic compounds, respec-
tively. For the O1s spectrum (Fig. 9b), two peaks are found at
Fig. 13 The mechanical properties of PP0, PP1, PP2, PP6 and PP9.
Fig. 14 TGA-FTIR spectra of pyrolysis products during the thermal degradation of PP6.
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 88445–88455 | 88453

View Article Online
RSC Advances
Paper
Fig. 15 (a) C1s and (b) O1s XPS spectra of the residue of PP6.
around 532.5 and 533.5 eV. It is impossible to distinguish
inorganic and organic oxygen because the O1s band is struc-
tureless. The peaks at 532.5 and 533.5 eV are assigned to ]O in
carbonyl or phosphate groups and –O– in C–O–C, C–O–P, or C–
OH groups, respectively. The char in form of these structures for
PP6 can effectively prevent the substrate materials from
burning, thus leading to an efficient ame retardancy.
2 Y. Liu, C. L. Deng, J. Zhao, J. S. Wang, L. Chen and
Y. Z. Wang, An efficiently halogen-free ame-retardant
long-glass-ber-reinforced polypropylene system, Polym.
Degrad. Stab., 2011, 96, 363–370.
3 H. Yu, Z. Jiang, J. W. Gilman, T. Kashiwagi, J. Liu, R. Song
and T. Tang, Promoting carbonization of polypropylene
during combustion through synergistic catalysis of a trace
of halogenated compounds and Ni2O3 for improving ame
retardancy, Polymer, 2009, 50, 6252–6258.
4. Conclusions
4 S. Bourbigot, M. le Bras, S. Duquesne and M. Rochery,
Recent advances for intumescent polymers, Macromol.
Mater. Eng., 2004, 289, 499–511.
A
microencapsulated ammonium polyphosphate with
melamine-formaldehyde-tris(2-hydroxyethyl)isocyanurate resin
was successfully prepared by in situ polymerization. The water
resistance and thermal stability of MFT–APP was improved
greatly compared with APP. When 30 wt% MFT–APP was added
into PP, UL-94 V-0 rating was achieved with a LOI value of 32%.
To further improve the ame retardancy, THEIC was added into
the PP/MFT–APP composites, a remarkable improvement of
ame retardation could be found. The maximum LOI values
increased to 36% and an excellent swell char layer was formed.
The results of cone calorimeter test indicated that the addition
of MFT–APP or MFT–APP/THEIC signicantly decreased the
HRR, pHRR, THR, pSPR and TSP. The results of water resistance
test illustrated that MFT–APP was benecial for improving the
ame durability of PP/IFR composites. The TGA-FTIR and XPS
results indicated that PP/MPP/TBM could form a compact and
thermostable char residue, which could effectively protect the
substrate material from burning. All the results illustrated that
microencapsulating APP with this MFT-resin was a promising
way to endow the PP/IFR with good water resistance and ame
retardancy.
5 N. Tian, X. Wen, Z. Jiang, J. Gong, Y. Wang, J. Xue and
T. Tang, Synergistic Effect between a Novel Char Forming
Agent and Ammonium Polyphosphate on Flame
Retardancy and Thermal Properties of Polypropylene, Ind.
Eng. Chem. Res., 2013, 52, 10905–10915.
6 G. Camino, L. Costa and L. Trossarelli, Study of the
mechanism of intumescence in re retardant polymers:
part III—effect of urea on the ammonium polyphosphate–
pentaerythritol system, Polym. Degrad. Stab., 1984, 7, 221–
229.
7 G. Camino, L. Costa, L. Trossarelli, F. Costanzi and
A. Pagliari, Study of the mechanism of intumescence in re
retardant polymers: part VI—mechanism of ester
formation in ammonium polyphosphate–pentaerythritol
mixtures, Polym. Degrad. Stab., 1985, 12, 213–228.
8 G. Camino, L. Costa and G. Martinasso, Intumescent re-
retardant systems, Polym. Degrad. Stab., 1989, 23, 359–376.
9 Y. Chen and Q. Wang, Reaction of melamine phosphate with
pentaerythritol and its products for ame retardation of
polypropylene, Polym. Adv. Technol., 2007, 18, 587–600.
10 G. Camino, L. Costa and L. Trossarelli, Study of the
mechanism of intumescence in re retardant polymers:
part II—mechanism of action in polypropylene–
References
ammonium
polyphosphate–pentaerythritol
mixtures,
1 P. Song, Z. Fang, L. Tong, Y. Jin and F. Lu, Effects of metal
Polym. Degrad. Stab., 1984, 7, 25–31.
chelates on
a
novel oligomeric intumescent ame
11 G. Camino, L. Costa and L. Trossarelli, Study of the
retardant system for polypropylene, J. Anal. Appl. Pyrolysis,
2008, 82, 286–291.
mechanism of intumescence in re retardant polymers:
88454 | RSC Adv., 2015, 5, 88445–88455
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
part I—thermal degradation of ammonium polyphosphate– 21 K. Wu, L. Song, Z. Wang and Y. Hu, Preparation and
pentaerythritol mixtures, Polym. Degrad. Stab., 1984, 6, 243–
252.
12 G. Camino, L. Costa and L. Trossarelli, Study of the
characterization of double shell microencapsulated
ammonium polyphosphate and its ame retardance in
polypropylene, J. Polym. Res., 2008, 16, 283–294.
mechanism of intumescence in re retardant polymers: 22 L. G. Wei, M. L. Xie and X. X. Chai, Synthesis of the
part V—mechanism of formation of gaseous products in
the thermal degradation of ammonium polyphosphate,
Polym. Degrad. Stab., 1985, 12, 203–211.
cocondensation resin of THEIC, melamine and phenol-
formaldehyde and its application, Eng. Plast. Appl., 1989, 4,
10–13.
13 M. Chen, Y. Xu, X. Chen, Y. Ma, W. He, J. Yu and Z. Zhang, 23 W. R. Li, H. L. Zhang, X. P. Liu and H. J. Song, Foam level
Thermal stability and combustion behavior of ame-
special
tris(2-hydroxyethyl)isocyanurate
modied
retardant polypropylene with thermoplastic polyurethane-
melamine resin, Thermosetting Resin, 2011, 26, 32–35.
microencapsulated ammonium polyphosphate, High 24 H. L. Zhang, X. P. Liu, J. W. Zhang and H. J. Song, Study on
Perform. Polym., 2014, 26(4), 445.
14 C. L. Deng, S. L. Du, J. Zhao, Z. Q. Shen, C. Deng and
foam level THEIC modied melamine resin, Chemical
Intermediate, 2011, 1, 28–30.
Y. Z. Wang, An intumescent ame retardant polypropylene 25 M. Jimenez, S. Duquesne and S. Bourbigot, Intumescent re
system with simultaneously improved ame retardancy
and water resistance, Polym. Degrad. Stab., 2014, 108, 97–
107.
protective coating: toward a better understanding of their
mechanism of action, Thermochim. Acta, 2006, 449, 16–26.
26 W. Y. Chen, S. S. Yuan, Y. Sheng and G. S. Liu, Effect of
charring agent THEIC on ame retardant properties of
polypropylene, J. Appl. Polym. Sci., 2015, 132, 33–40.
27 S. S. Yuan, W. Y. Chen and G. S. Liu, Synergistic effect of
THEIC-based charring agent on ame retardant properties
of polylactide, J. Appl. Polym. Sci., 2015, 132, 65–69.
15 K. Cao, S. L. Wu, K. L. Wang and Z. Yao, Kinetic Study on
Surface Modication of Ammonium Polyphosphate with
Melamine, Ind. Eng. Chem. Res., 2011, 50, 8402–8406.
16 Z. B. Shao, C. Deng, Y. Tan, M. J. Chen, L. Chen and
Y. Z. Wang, Flame retardation of polypropylene via a novel
intumescent ame retardant: ethylenediamine-modied 28 W. Y. Chen and G. S. Liu, Flame-retardancy properties of
ammonium polyphosphate, Polym. Degrad. Stab., 2014,
106, 88–96.
tris(2-hydroxyethyl)isocyanurate based charring agents on
polypropylene, J. Appl. Polym. Sci., 2015, 132(16), 5126–5132.
17 H. Lin, H. Yan, B. Liu, L. Wei and B. Xu, The inuence of KH- 29 B. Du, Z. Guo, P. A. Song, H. Liu, Z. Fang and Y. Wu, Flame
550 on properties of ammonium polyphosphate and
polypropylene ame retardant composites, Polym. Degrad.
Stab., 2011, 96, 1382–1388.
retardant mechanism of organo-bentonite in polypropylene,
Appl. Clay Sci., 2009, 45, 178–184.
30 X. Su, Y. Yi, J. Tao and H. Qi, Synergistic effect of zinc
hydroxystannate with intumescent ame-retardants on re
retardancy and thermal behavior of polypropylene, Polym.
Degrad. Stab., 2012, 97, 2128–2135.
18 Z. Lei, Y. Cao, F. Xie and H. Ren, Study on surface
modication and ame retardants properties of
ammonium polyphosphate for polypropylene, J. Appl.
Polym. Sci., 2012, 124(1), 781–788.
31 A. B. Morgan and M. Bundy, Cone calorimeter analysis of
UL-94 V-rated plastics, Fire Mater., 2007, 31, 257–283.
19 K. Wu, Z. Wang and H. Liang, Microencapsulation of
ammonium polyphosphate: preparation, characterization, 32 B. Schartel and T. R. Hull, Development of re-retarded
and its ame retardance in polypropylene, Polym. Compos.,
2008, 29, 854–860.
materials interpretation of cone calorimeter data, Fire
Mater., 2007, 31, 327–354.
20 L. Yang, W. Cheng, J. Zhou, H. Li, X. Wang, X. Chen and 33 P. J. Elliot and R. H. Whiteley, A cone calorimeter test for the
Z. Zhang, Effects of microencapsulated APP-II on the
microstructure and ame retardancy of PP/APP-II/PER
composites, Polym. Degrad. Stab., 2014, 105, 150–159.
measurement of ammability properties of insulated wire,
Polym. Degrad. Stab., 1999, 64, 577–584.
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 88445–88455 | 88455