Synthesis and performance of star-shaped aluminum phosphinate flame retardant

Article abstract of DOI:10.1007/s10973-015-5222-8

A star-shaped aluminum dimethyl(2,2-dimethyl-1,3-propylene)bis(oxy)bicarbonylethyldiphos phinate (AlCP) was synthesized by esterifying reaction between 2-methy-2,5-dioxo-1,2-oxaphospholane (OP) and neopentyl glycol following by neutralizing with aluminum hydroxide. The structure and composition were confirmed with FTIR, ¹H NMR, ³¹P NMR and X-ray fluorescent spectroscopy. AlCP begins to decompose at 370?°C, matching the initial decomposition temperature of PBT. It endows PBT with desired flame retardancy and anti-dripping property and shows good compatibility with PBT. When the total filler loading is kept at 17, 15–13?mass% AlCP combining with 2–4?mass% MC can make the PBT passing the UL 94 V0 rating and eliminating dripping. In addition, the AlCP has a little influence on the mechanical properties of PBT. The tensile strength and notched impact strength of PBT containing 17?mass% AlCP only reduce by 5.6 and 4?%, respectively, comparing to the pure PBT. The SEM photographs show that AlCP has a good compatibility with PBT. Microcombustion calorimeter, TG-FTIR and SEM results show that AlCP can depress the heating release and promote char forming during combustion and acts mainly in the condensed phase.

Full text of DOI:10.1007/s10973-015-5222-8

J Therm Anal Calorim (2016) 124:1399–1409
DOI 10.1007/s10973-015-5222-8
Synthesis and performance of star-shaped aluminum phosphinate
flame retardant
Liyong Zou1,2 Jiyan Liu1,2 Xueqing Liu1,2 Xiaomeng Wang1,2
Jia Chen1,2
•
•
•
•
Received: 23 July 2015 / Accepted: 22 December 2015 / Published online: 23 February 2016
´
´
Ó Akademiai Kiado, Budapest, Hungary 2016
Abstract A star-shaped aluminum dimethyl(2,2-dime-
thyl-1,3-propylene)bis(oxy)bicarbonylethyldiphos phinate
(AlCP) was synthesized by esterifying reaction between
2-methy-2,5-dioxo-1,2-oxaphospholane (OP) and neo-
pentyl glycol following by neutralizing with aluminum
hydroxide. The structure and composition were confirmed
with FTIR, 1H NMR, 31P NMR and X-ray fluorescent
spectroscopy. AlCP begins to decompose at 370 °C,
matching the initial decomposition temperature of PBT. It
endows PBT with desired flame retardancy and anti-drip-
ping property and shows good compatibility with PBT.
When the total filler loading is kept at 17, 15–13 mass%
AlCP combining with 2–4 mass% MC can make the PBT
passing the UL 94 V0 rating and eliminating dripping. In
addition, the AlCP has a little influence on the mechanical
properties of PBT. The tensile strength and notched impact
strength of PBT containing 17 mass% AlCP only reduce by
5.6 and 4 %, respectively, comparing to the pure PBT. The
SEM photographs show that AlCP has a good compatibility
with PBT. Microcombustion calorimeter, TG-FTIR and
SEM results show that AlCP can depress the heating
release and promote char forming during combustion and
acts mainly in the condensed phase.
Keywords PBT Á Flame retardancy Á Metal phosphinate Á
Star-shaped phosphinate Á Polyester
Introduction
The poly(butyleneterephthalate) (PBT) is widely used as
engineering plastic in electrical, electronic and automotive
components. PBT is inherent flammability and serious
dripping during combustion and involves a real hazard to
the users for these applications. Hence, improving its fire
resistance is very important [1, 2]. The halogen compounds
are commonly used flame retardants for PBT; however,
they have become under scrutiny because of ecological
concerns [3, 4]. At this time, the flame retardant industry is
under pressure to change to flame retardants that are per-
ceived to be more environmentally friendly and harmless to
health [5, 6].
Metal salts of phosphinate belong to a new-generation
environment-friendly flame retardants. The roles of these salts
in the improvement of flammability properties and in the
influence of mechanical properties and the thermal stability of
the composites are related to the structure of phosphinate
anion and metal type of the salts [7–13]. It was found that the
aluminumorcalcium ethylmethylphosphinategivesV0rating
in the UL 94 test at 15 mass% loading in PBT and 20 mass%
loading in glass-filled PBT, while the sodium ethyl-
methylphosphinate gives V0 rating at 30 mass% of filler
loading in PBT [14]. In the case of aluminum salts of phos-
phinate-based PBT, aluminum 1-(methoxyethyl)methylphos-
phinate, aluminum 1-(ethoxyethyl)methylphosphinate and
aluminum 1-(methoxyethyl)ethylphosphinate [15] showed
only V1 rating at 20 mass% loading in glass-filled PBT.
Aluminum hydroxymethylmethylphosphinate, despite of its
highphosphoruscontent, showedonlyV2ratinginglass-filled
& Jiyan Liu
liuxueqing2000@163.com
1
Key Laboratory of Optoelectronic Chemical Materials and
Devices of Ministry of Education, Jianghan University,
Wuhan 430056, China
2
Flexible Display Materials and Technology Co-innovation
Center of Hubei Province, Jianghan University,
Wuhan 430056, China
123

1400
L. Zou et al.
PBT [16], while aluminum methylethylphosphinate or alu-
minum methylcyclohexylphosphinate shows V1 rating in
PBT at 17 mass% filler loading. In our previous research [12,
17], influence of alkyl group of the aluminum salts of phos-
phinates on the properties of EP has been investigated. It was
found that the metal type has more influence on the flame
retardancy of the polymer, while the alkyl groups of the
phosphinates affect much more the mechanical properties of
the polymer and dispersion of the particles in matrix.
forming agent in the flame retardant system [32–34]; two
hydroxyl groups in neopentyl glycol can react with OP
forming a star-shaped molecule. When Al(CP) serves as a
flame retardant in polymer such as PBT, the bulk organic
ester group of the AlCP is expected to present high carbon
forming ability and good compatibility with polymer.
Experimental
Materials
Commercial metal salts of phosphinates formulas for
PBT are based on aluminum or zinc diethylphosphinate.
However, the damage of these salts on the mechanical
properties of the final materials should not be ignored.
Poly(butyleneterephthalate) (PBT) was obtained from
Nantong Xincheng materiel Ltd. (Jiangsu, China). Mela-
mine Cyanurate (MC) was purchased from Taixing Fine
chemical Ltd. (Jinan, China). 2-Methy-2,5-dioxo-1,2-
oxaphospholane (OP) with purity above 98 % was pur-
chased by Zhenghao Chemical Ltd. (Wuhan, China).
Neopentyl glycol, acetone, toluene, p-toluenesulfonic acid
and aluminum isopropoxide are chemical reagents and
purchased from Guoyao Chemical Ltd. (Tianjin, China)
and used as received.
¨
Koppl [18] investigated the effect of aluminum salts on
the mechanical and thermal properties of PBT and glass
fiber-reinforced PBT. For the case of PBT, 20 mass% of
aluminum salts is the minimum dosage for PBT to pass
the V0 rating; however, at this dosage, the tensile strength
of the material is reduced by about 35 %, unnotched
impact strength is reduced significantly from 190 to
20 kJ m-2 and the elongation at break for ductility is
reduced by 76 %. The deterioration of mechanical prop-
erties of the material was due to the bad dispersion of the
salt particles in PBT and inferior compatibility between
particle and PBT.
Synthesis of b-carboxylethylmethylphosphinic acid
(CEP)
Balance among various properties such as thermal sta-
bility, compatibility, mechanical properties and flame
retardancy is very important for high-performance flame
retardants/polymer system [19, 20]. The phosphinate salts
possessing the properties such as high flame retardancy,
anti-dripping during combustion and good compatibility
with matrix are therefore highly desired [21–24]. In addi-
tion, a highly efficient flame retardant should show a
decomposition temperature similar to that of the polymer.
Matching of the thermal degradation temperature between
the flame retardant and polymer can make the flame
retardant to act the best in both gas phase and condensed
phase.
To a solution of 0.2 mol (26.8 g) OP in 100 mL of acetone
was added dropwise 0.2 mol (3.60 g) water within 30 min
at 50 °C. After the addition was completed, the mixture
was refluxing for 2 h to obtain CEP. CEP was washed with
acetone thrice and then dried at 100 °C for 24 h (yield
93 %).
Synthesis of dimethyl(2,2-dimethyl-1,3-
propylene)bis(oxy)bicarbonylethyldiphosphinic acid
(CP)
0.11 mol CEP (16.72 g), 0.05 mol (5.21 g) neopentyl
glycol and 0.01 mol (0.17 g) p-toluenesulfonic acid were
placed in a round-bottom flask equipped with a temperature
controller, a reflux condenser with side arm, a nitrogen pipe
and a mechanical stirrer. The mixture was heated at 120 °C
under refluxing for 4 h. The white product was collected,
washed with acetone (3 9 100 mL) and dried in air to
obtain the CP (yield 94 %).
In order to improve the fire resistant efficiency and other
properties of the metal salts of phosphinate-based polymer
materials, adding nanoparticles or nitrogen-containing
compounds such as melamine compounds has been tried
[25–28]. Designing metal salts of phosphinate with
appropriate structure has also been a good solution [29–31].
This article focuses on a star-shaped aluminum dime-
thyl(2,2-dimethyl-1.3-propylene)bis(oxy)bicar-
bonylethyldiphosphinate(AlCP). It was synthesized from
commercial available materials 2-methy-2,5-dioxo-1,2-
oxaphospholane (OP), neopentyl glycol and aluminum iso-
propoxide using a gentle and easy approach, as shown in
Fig. 1:
Synthesis of aluminum salt of dimethyl(2,2-
dimethyl-1,3-propylene)bis(oxy)bicarbonyl
ethyldiphosphinate (AlCP)
To a solution of 0.15 mol (55.8 g) CP in 200 mL of iso-
propanol, a solution of 0.1 mol (20.5 g) aluminum iso-
propoxide in 50 mL of isopropanol was added. The
Choosing neopentylglycol as reactant with OP is based
on the following consideration: Polyol is often used as char
123

Synthesis and performance of star-shaped aluminum phosphinate flame retardant
1401
CH3
O
C
CH3
CH2 CH2
OH
O
O
P
O
O
P
P
p-toluenesulfonic acid/toluene
130 °C, 4 h
H2C
H2C
H2O
H2C
H2C
O
H3C
H3C
H3C
OH
O
H3C
HO CH2
CH3
CH2
CH3
P
C
O
(CH2)2
COOH
OH
C
O
OH
CH2 CH2
O
OP
CEP
CP
H3C
CH3
O
O
C
CH2
CH2
CH2
O
O
O
C
O
P
CH3
CH2
CH3
P
O
C
O
O
CH3
P
Al
OH
CH2 CH2
O
CH3
O
H3C
CH3
H2C
O
O
O
H3C
H3C
P
+
isopropanol, 85 °C, 3 h
Al
CH3
CH3
O
O
CH2
CH2
O
O
P
CH3
H3C
O
CH3
P
CH3
CH2
O
O
O
H2C
Al
CH3
C
C
O
OH
P
C
O
CH2
CH2
O
O
O
O
H2C
CH2
CH2
O
P
CH3
H2C
CH3
C
O
H3C
H3C
O
C
CH3
CH2
O
O
CP
AICP
Fig. 1 Synthetic route for AlCP
mixture was stirred at 85 °C under refluxing for 3 h. The
crude white precipitate AlCP was washed with alcohol and
dried at 80 °C for 24 h (yield 96 %).
Table 1 Formulation and flame retardancy of the samples
Sample
Formulation/mass% UL94 LOI/% Dripping
PBT AlCP MC
Preparation of composites
PBT
10
87
0
0
0
–
16.6
22.9
25.7
20.6
23.5
26.8
28.3
26.4
22.5
Yes
Yes
Yes
Yes
No
PBT/AlCP1
13
10.5
9.0
17
15
13
11
0
V-2
All the samples were prepared by melt blending using an
XK-160 twin-screw internal mixture (Jiangsu, China) at
235 °C for 20 min with a screw speed of 120 rpm. The
resulting mixture was then hot-pressed under 10 MPa at
240 °C for 5 min into sheet with suitable dimension for the
further measurements. The examined formulations are lis-
ted in Table 1.
PBT/AlCP/MC1 87
PBT/AlCP/MC2 87
2. 5 V-1
4
V-2
V-1
V-0
V-0
V-0
V-1
PBT/AlCP2
83
0
PBT/AlCP/MC3 83
PBT/AlCP/MC4 83
PBT/AlCP/MC5 83
2
No
4
No
6
Yes
Yes
PBT/MC
83
17
Measurements
analyzer. Samples of about 10 mg were heated in alumina
pans from room temperature up to 700 °C at a heating rate
of 20 °C min-1 in N2. TG was coupled with a Nicolet iS10
FTIR spectrometer (Perkin-Elmer, USA). Limiting oxygen
index (LOI) measurements were taken using a HC-2-type
instrument (Jiangning, China) in accordance with ASTM
D2863-77. The UL 94 vertical burning tests were con-
ducted on a CZF-3 instrument (Jiangning, China) in
accordance with ASTM D3801. The dimensions of the
sample were 100 9 13 9 3 mm3. The heat release rate
(HRR) and the total heat release (THR) were measured in a
1
The H NMR and 31P NMR spectra have been recorded in
a solution of DMSO-d6 at 25 °C with a Mercury VX-300
instrument operating at 400 MHz (Varian, US), using TMS
as inner reference and H3PO4 (85 %) as external reference.
FTIR was recorded on a Tensor 27 Bruker spectrometer
(Bruker, German) with KBr powder. The X-ray fluorescent
spectroscopy (XRF) measurement was done with a ZSX
Primus II (Rigaku, Japan) XRF spectrometer with a 35-kV
Rh-anode tube. Thermogravimetric analysis (TG) was
carried out with NETZSCH TG 209 thermogravimetric
123

1402
L. Zou et al.
MCC-2 microcombustion calorimeter (MCC) (Govmark
Organization Inc., USA). Samples of about 5–7 mg were
heated in alumina pans from 40 to 700 °C at a heating rate
of 1 °C s-1. The flow rate of N2 and O2 was 80 mL min-1
and 20 mL min-1, respectively. Samples were coated with
platinum in a vacuum. Tensile property was carried out
using a universal testing machine (CMT6000, SANS,
Shenzhen, China) following the standard ASTM D638. The
testing speed was kept at 5 mm min-1. Impact testing was
performed with an impact tester (ZBC1400-1, SANS,
Shenzhen, China) according to standard ASTM D256 Izod
Impact method. The fracture surfaces of specimens sub-
jected to tensile were gold-coated and examined by scan-
ning electron microscopy (SEM) using a Philips XL-40
instrument.
and 818 cm-1 (CH3) [35]. After CEP esterifying with
neopentyl glycol, in the spectrum of CP, the peaks around
3000 cm-1 become narrow, and peaks between 2700 and
2500 cm-1 disappear; in addition, the peak for C=O of acid
appearing at 1721 cm-1 shifts to 1731 cm-1, and peak for
C–O in COOH shifts to 1235 cm-1; above results indicate
–COOH has taken part in the reaction and ester group was
formed [36]. Other new peaks coming from neopentyl
glycol are 1472 cm-1 (–CH2) and 1377 cm-1 (C–(CH3)2)
in the spectrum of CP. In the spectrum of AlCP, the
absorption peak at the 987 cm-1 becomes weaker due to
the formation of P–O–Al.
1
Figure 3 shows H NMR and 31P NMR spectrum of
1
CEP, CP and AlCP. In the H NMR spectrum of CEP, the
shifts at 1.28, 1.32 ppm are for P–CH3, shifts at around
1.80 ppm are for P–CH2 and shifts at 2.36–2.43 ppm are
for C–CH2; the H in the COOH should be at 10 ppm, and
the H in the POH should be at about 5–7; however, only a
wide shift is observed with the peak value of 7.99 ppm. For
the molecule containing both OH and COOH, only peak
being observed is due to the exchange of H between OH
and COOH, the location of the shift depends on inter-
molecular or intramolecular hydrogen [36, 37], so in the
CEP, the shift at 7.93 ppm is attributed to the COOH and
Results and discussion
Chemical structure of AlCP
Figure 2 shows the FTIR spectra of CEP, CP and AlCP. In
the CEP, the OH absorption of –COOH is about
3000 cm-1, which overlaps with the absorption of CH3
(2964, 2911 cm-1). The broad peaks between 2700 and
2500 cm-1 are attributed to the COOH. The peak at
1720 cm-1 is for C=O and 1220 cm-1 is for C–O in
COOH. Other peaks are 1418 cm-1 (CH2), 1377 cm-1 (P–
CH3), 1157 cm-1 (P=O), 987 cm-1 (P–O–H), 884 cm-1
1
OH. In the H NMR of CP, new shifts at 0.88–90 and
3.82–3.84 ppm are due to the –CH3 and –CH2– in neo-
pentyl, respectively; the shift at 5.64 ppm is for the P–OH.
Comparing the 1H NMR of the AlCP with CP, the shift for
the POH disappears in the AlCP, indicating the OH has
Fig. 2 FTIR spectra for CEP,
CP and AlCP
1296
1422
1476
AICP
813
2964
760
1377
882
1734
1233
1079
1157
CP
2964
1314
1472
1377
756
816
1411
1050
1239
1731
1720
881
983
CEP
1126
2545
2290
2919
2998
634
733
1411
1314
1220
1122
813
988
888
1050
941
3500
3000
2500
2000
1500
1000
500
Wavenumber/cm–1
123

Synthesis and performance of star-shaped aluminum phosphinate flame retardant
1403
46.5
CEP
CEP
10
9
8
7
6
5
4
3
2
1
0
10
20
30
40
50
60
70
47.7
CP
CP
10
9
8
7
6
5
4
3
2
1
0
10
20
30
40
50
60
70
45.5
AICP
AICP
10
9
8
7
6
5
4
3
2
1
0
10
20
30
40
50
60
70
Chemical shift/ppm
Fig. 3 1H NMR and 31P NMR spectra for CEP and CP
Chemical shift/ppm
123

1404
L. Zou et al.
formed O–Al. In the 31P NMR, the shift for P of the CEP is
46.5 ppm, the shift for P of the CP is 47.7 ppm and the
shift for P of the AlCP is 45.5 ppm. The difference shift of
the P in three substances is due to the environment of the P
atom in the molecule. When CEP reacts with neopentyl
glycol, forming electron-rich ester CP causes phosphorus
to shift upfield. In the AlCP, the Al–O–P bond shows more
covalent properties and the electron density on the P atom
is less than that of P–OH, so the shift for the P moves to the
lower filed again. This phenomenon is consistent with the
explanation in the literature [37]. The XRF test shows that
the Al is 3.75 mass% and the P is 12.39 mass% in the
AlCP; the atomic ratio of P to Al is 2.87:1, closing to the
calculated value of 3:1.
1000
800
600
400
200
0
PBT
PBT/AICP2
PBT/AICP/MC3
PBT/AICP/MC4
PBT/AICP/MC5
360
380
400
420
440
460
480
Temperature/°C
The FTIR, 1H NMR, 31P NMR and the XRF data for the
products are consistent with the structure as shown in
Fig. 1.
Fig. 4 HRR curves for PBT, PBT/AlCP2 and PBT/AlCP/MC
formulations
HRC values. MCC results are consistent with the LOI and
UL 94 performance.
Combustion characteristics: LOI, UL 94 and MCC
Table 1 reports the UL 94 test and LOI value of the sam-
ples with varying formulation. PBT is highly flammable
with the LOI of 16 % and fails in the UL 94 test burning
with flammable dripping. Adding 17 mass% AlCP makes
PBT passing the V1 rating with the LOI increasing to
23.5 % and no dripping during combustion. When the total
filler loading is kept at 17 mass% and AlCP is partially
substituted with MC, PBT/AlCP/MC3 and PBT/AlCP/
MC4 show satisfying flame retardancy and pass the V0
ratio without dripping during combustion. PBT/AlCP/MC5
passes the V0 rating but dripping during combustion. PBT/
MC just passes the V1 rating with LOI of 22.5 % and
dripping. Above results indicate that AlCP combining with
MC in an appropriate ratio shows a good flame retardancy
and anti-dripping property.
Char residues analysis
Figure 5 illustrates the SEM photographs of the char resi-
dues after combustion. The char surface of PBT/AlCP2 is
smooth, and a few holes of diameter less than 5 lm dis-
tribute on it. For the PBT/AlCP/MC formulations, the char
surface becomes rough, more and bigger holes are observed,
and especially the char for PBT/AlCP/MC4 has the biggest
hole size. The number and the hole size on the char surface
is in the order of PBT/AlCP/MC4[ PBT/AlCP/MC3 [
PBT/AlCP/MC5 [ PBT/AlCP2, implying the most volatile
releasing from the PBT/AlCP/MC4 and the least volatile
coming from the PBT/AlCP2 during the combustion.
The MC mainly acts in the gas phase by releasing the
nonflammable gas in the PBT [35, 38]. It is supposed that
the AlCP acts mainly in the condensed phase by forming
the dense carbonized char. For the PBT/AlCP2, most of the
AlCP takes part in the char forming and only a little
volatile produced from the AlCP during the combustion;
hence, the surface of the char is smooth, and a few and
small holes are observed. When MC instead of the AlCP is
incorporated in the PBT/AlCP, MC decomposes into gas
coming out of the char layer and leaving lots of big holes
on the surface. With the ratio of the MC/AlCP increasing,
the char layer becomes loose, more volatile producing from
MC easily dashes out of the char layer and so bigger holes
and rougher surface are observed. With the MC increasing
continually, in the case of PBT/AlCP/MC5, the number
and the size of the holes do not increase correspondingly,
and the char surface looks smoother than the char surface
from PBT/AlCP/MC4. Smaller hole size and smoother
surface indicate less volatile dashing out of the char
PBT, PBT/AlCP2 as well as the PBT/AlCP/MC for-
mulations having passed the V0 rating were further
investigated with MCC. The heat release rate (HRR) curves
are shown in Fig. 4, and the corresponding data such as the
peak of the heat release rate (PHRR), the temperature for
PHRR (TPHRR), the heat release capacity (HRC) and the
total heat release (THR) are presented in Table 2. Adding
AlCP reduces obviously the PHRR, THR and HRC of the
PBT. When the AlCP was partially replaced with MC, in
the case of PBT/AlCP/MC formulations, the TPHRR and
PHRR increase, whereas the HRC and THR decrease
comparing to PBT/AlCP2. It implies that AlCP combining
with MC can move the combustion of PBT to a higher
temperature. The HRC is a good predictor of the flamma-
bility, and it reflects a true material property rooting in the
structure and composition of the materials. Among the
three PBT/AlCP/MC formulations, the PBT/AlCP/MC4
shows the highest TPHRR and the lowest PHRR, THR and
123

Synthesis and performance of star-shaped aluminum phosphinate flame retardant
1405
Table 2 Data from MCC experiments
Samples
TPHRR/°C
PHRR/W g-1
THR/kJ g-1
HRC/J g-1 K-1
PBT
426
425
428
430
427
2
3
2
2
3
889 14
706 16
795 11
762 18
838 20
23.3 0.1
19.0 0.2
19.4 0.2
19.3 0.1
19.9 0.2
661 10
533 20
557 14
547 17
589 19
PBT/AlCP2
PBT/AlCP/MC3
PBT/AlCP/MC4
PBT/AlCP/MC5
Fig. 5 SEM results for a PBT/AlCP2, b PBT/AlCP/MC3, c PBT/AlCP/MC4 and d PBT/AlCP/MC5 after burning
surface. In order to investigate this phenomenon, FTIR
coupling with thermal degradation analysis was used to
monitor the volatile releasing.
improved. When the total filler loading is kept at
17 mass%, the AlCP is partially replaced with MC, the
T5 %, Tmax and char at 700 °C of the PBT/AlCP/MC for-
mulations are reduced comparing with PBT/AlCP. During
being heated, the MC decomposed into gas, and most of the
AlCP forms the char residues remaining in the condensed
phase. Theoretically, the char yield should reduce with MC
increasing. However, TG results show that the PBT/AlCP/
MC4 has the lowest char yield among the three PBT/AlCP/
MC formulations. The probable reason is following: MC
decomposes earlier than AlCP, at higher MC content, the
lots of nonflammable gas release from the MC before the
decomposition of the matrix and AlCP, this nonflammable
gas depressed decomposition of the components in for-
mulation, some of the components that did not decompose
Thermal degradation analysis
Figure 6 shows the TG-DTG curves of samples with dif-
ferent formulation in N2. Detailed and precise parameters
including the temperature at 5 % mass loss (T5 %), the
maximum rate degradation temperature (Tmax), the maxi-
mum decomposition rate (DTGmax) and the char yields at
700 °C are summarized in Table 3. AlCP shows good
thermal stability with the T5 % 368 °C, and the char at
700 °C is 35.3 %. Adding AlCP reduces the T5 % and
DTGmax of PBT slightly, and the char residues of PBT are
123

1406
L. Zou et al.
100
80
60
40
20
0
0.08
0.06
0.04
0.02
0.00
AICP
PBT
PBT/AICP2
PBT/AICP/MC3
PBT/AICP/MC4
PBT/AICP/MC5
PBT
PBT/AICP2
PBT/AICP/MC3
PBT/AICP/MC4
PBT/AICP/MC5
100
200
300
400
500
600
700
5
10
15
20
25
30
35
Temperature/°C
Time/min
Fig. 7 Intensity of the total pyrolysis gas for the samples at different
times
0
–10
–20
–30
–40
–50
–60
0.4714. Adding AlCP into PBT reduces the intensity of the
volatile greatly, and the total area of the peaks for the PBT/
AlCP2 is 0.2373. After the AlCP was partially replaced with
MC, for the PBT/AlCP/MC formulations, the peak of the
volatile becomes wider and two peaks are observed. The
MC starts to decompose at about 310 °C, so the first peak is
mainly contributed by the MC, and the second peak is
contributed by all the components in the formulation. In
addition, the peaks for the PBT/AlCP/MC4 appear earlier
and the total area of the peaks (0.3607) is larger than that of
PBT/AlCP/MC3(0.2380) or PBT/AlCP/MC5 (0.2487),
indicating PBT/AlCP/MC4 has lower decomposition tem-
perature and produces more volatile than PBT/AlCP/MC3
and PBT/AlCP/MC5, which agrees with the SEM observa-
tion in Fig. 5 that lots of gas releasing from the sample lead
to the large holes in the char surface of PBT/AlCP/MC4.
FTIR spectra for the volatile coming from the various
formulations at 17, 19 and 22 min are presented in Fig. 8.
At 17 min, the temperature is about 380 °C, the pure PBT
just starts degrading and almost no signal is detected.
When the AlCP is added into PBT, signals for the volatile
coming from PBT in the case of PBT/AlCP2 can be
detected as following: 3730 and 3587 cm-1 (benzoic acid),
3587 cm-1 (water), 3038 cm-1 (benzene), 2970 and
1021 cm-1 (tetrahydrofuran THF) 2350 and 617 cm-1
(CO2), 1741, 1268 and 1108 cm-1 (ester), 915 cm-1 (bu-
tadiene) and 734 cm-1 (C–H in aromatic ring) [8, 12]. For
the PBT/AlCP/MC formulations, the signals at 2350 and
2380 cm-1 become intense and a new signal for the HOCN
deriving from the MC is observed at 2252 cm-1. CO2 and
HOCN are main products during degradation of MC [19];
of course they are higher in concentration in PBT/AlCP/
MC than in PBT or PBT/AlCP. At 19 min, the PBT
reaches the maximum degradation rate; strong signals for
the volatile from neat PBT can be observed. Adding AlCP
into PBT leads that all signals relating to PBT become
AICP
PBT
PBT/AICP2
PBT/AICP/MC3
PBT/AICP/MC4
PBT/AICP/MC5
300
350
400
450
500
550
Temperature/°C
Fig. 6 TG-DTG curves for the samples in N2
Table 3 TG-DTG data for the samples in N2
Sample
T5 %/°C
Tmax/°C
DTGmax
% min-1
/
Char at
700 °C/%
AlCP
369
377
374
362
347
354
430
410
411
410
403
408
42.9
55.2
46.0
46.6
46.8
49.7
35.3
4.4
9.9
8.7
6.4
7.0
PBT
PBT/AlCP2
PBT/AlCP/MC3
PBT/AlCP/MC4
PBT/AlCP/MC5
then were trapped by the char, so the char yield increases
again in the PBT/AlCP/MC5.
FTIR analysis for volatilized products
Figure 7 presents the intensity of total volatile with time for
the various formulations. The area under the curve is the
total intensity of the peaks; it can be obtained by integrating
the curve. The pure PBT starts releasing volatile after
17 min, and one peak is observed with the area of the peaks
123

Synthesis and performance of star-shaped aluminum phosphinate flame retardant
1407
Table 4 Mechanical properties of the samples
a
b
Sample
ex/%
E/MPa
TS/MPa NIS/kJ m-2
PBT
19.6 1.1 494
8
53
52 12 5.8 0.3
5.7 0.4
PBT/AlCP/MC2 15.6 1.6 509 11 48 10 5.6 0.3
PBT/AlCP2 15.2 1.3 536 50 11 5.8 0.4
9
6.0 0.2
PBT/AlCP1
17.1 1.5 514
9
PBT/AlCP/MC1 16.2 1.7 523 10 50
8
c
9
PBT/AlCP/MC3 15.4 1.4 506 10 50 11 5.5 0.2
PBT/AlCP/MC4 14.9 1.6 514 12 50 12 5.4 0.3
d
PBT/AlCP/MC5 14.5 1.4 522 13 49
8
5.3 0.2
e
PBT/MC 14.3 1.2 533 11 49 10 5.1 0.1
17 min
4000
3000
3000
3000
2000
1000
Wavenumber/cm–1
1015 cm-1 due to the P=O and P–O of the AlCP are
observed in the formulations containing AlCP.
TG-FTIR results show that adding AlCP leads the PBT
to decompose faster in the early stage and then depress the
degradation of the PBT at higher temperature, which is
similar to the function of aluminum methylcyclo-
hexylphosphonates [30]. On the other hand, the volatile
coming from the AlCP is detected at higher temperature
and the signals are weaker than those producing by PBT
and MC, indicating most of AlCP was left in the condensed
phase to prevent the PBT from degradation.
a
b
c
d
e
MC mainly acts in the gas phase by releasing the non-
flammable gas diluting the flammable substance [39, 40].
Appropriate combination of AlCP and MC can make the
flame retardants acting in the gas and the condensed phase
simultaneously and shows optimum synergistic effect by
forming loose char layer to isolate the heat and flammable
gas from the inner layer of the materials. Combining flam-
mable test and MCC determination indicates that the PBT/
AlCP/MC4 shows the satisfying flame retardancy and anti-
dripping properties during combustion. SEM photographs
and TG-FTIR measurement show that PBT/AlCP/MC4
formed the loose char layer and released more volatile than
the other two PBT/AlCP/MC formulations. It needs a further
investigation in the future why the formulation with specific
combination shows the better synergism than others.
19 min
4000
2000
1000
Wavenumber/cm–1
a
b
c
d
e
Mechanical properties
22 min
The tensile properties including elongation at breaking (e %),
tensile strength (TS), tensile modulus (E) and notched impact
strength (NIS) are summarized in Table 4. Adding AlCP into
PBT reduces the e %, TS and NIS of the PBT, but the E is
increased. The reduction in above-mentioned parameters is
far less than the diethylphosphinate-filled PBT [18]. For the
PBT/AlCP2, the ex, TS and NIS values are lower by 22, 5.6
and 4 %, respectively, than pure PBT. When the AlCP is
partially replaced with MC, the mechanical properties of the
samples almost have no change.
4000
2000
1000
Wavenumber/cm–1
Fig. 8 FTIR spectra of pyrolysis products for a PBT, b PBT/AlCP2,
c PBT/AlCP/MC3, d PBT/AlCP/MC4, e PBT/AlCP/MC5 at different
times
weak except signals at 2970 and 3080 cm-1, where the
signals of THF overlap with the signals of CH2 from the
AlCP. At 22 min, new signals at the 1200, 1049 and
123

1408
L. Zou et al.
Fig. 9 SEM results for fracture surface of a PBT/AlCP2, b PBT/AlCP/MC3, c PBT/AlCP/MC4 and d PBT/AlCP/MC5 before burning
Figure 9 presents the SEM photographs for the fracture
surface of the samples. Almost no naked particles are seen
on the surface, indicating a good compatibility between
PBT and AlCP. Good compatibility is probably due to the
bulk ester-containing organic group in the AlCP, which has
a similar structure and polarity to the PBT.
When the total filler loading is kept at 17, 13–14 mass%
AlCP combining with 2–4 mass% MC can make the PBT
passing the V0 rating and eliminating dripping. In the PBT/
AlCP/MC formulations, PBT/AlCP/MC4 containing
4 mass% MC and 13 mass% AlCP shows the best flame
retardancy and lowest heat release during combustion. SEM
photographs show that the char from the PBT/AlCP/MC4 is
looser and rougher than other PBT/AlCP/MC formulations.
The reasons why the components in PBT/AlCP/MC4 have a
maximum synergistic effect need to investigate in the future
work.
Conclusions
Star-shaped aluminum phosphinates AlCP was success-
fully synthesized. The synthesis process and the products
were characterized by the FTIR, NMR and XRF. AlCP is
endowed with good thermal stability, desired flame retar-
dancy, anti-dripping property as well as good dispersion in
PBT. The AlCP begins to decompose at 370 °C in nitro-
gen, matching the initial decomposition temperature of
PBT very well. Similar decomposition temperature
between the flame retardant and polymer can make the
flame retardant to play a maximum role in both gas phase
and condensed phase during polymer combustion.
Acknowledgements This work is supported by of the academic
innovation program for the outstanding youth groups of universities in
Hubei (T201318) and key project of nature science founding of Hubei
(2014CFA098).
References
1. Levchik SV, Weil ED. A review on thermal decomposition and
combustion of thermoplastic polyesters. Polym Adv Technol.
2004;15:691–700.
Adding 17 mass% AlCP alone in PBT can make the
PBT reaching the UL94 V1 rating and eliminating drip-
ping. TG-FTIR results showed that the AlCP catalyzes the
decomposition of the PBT at the early decomposition stage
and promotes char forming at higher temperature. And
more importantly, the AlCP has a little influence on the
mechanical properties of the PBT. The tensile strength and
notched impact strength of the PBT containing 17 mass%
AlCP only reduce by 5.6 and 4 %, respectively, comparing
with the pure PBT. The SEM photographs show that AlCP
has a good compatibility with PBT.
2. Levchik SV, Weil ED. Flame retardancy of thermoplastic
polyesters—a review of the recent literature. Polym Int.
2005;54:11–35.
3. Veen IV, Boer J. Phosphorus flame retardants: properties, pro-
duction, environmental occurrence, toxicity and analysis. Che-
mosphere. 2012;88:1119–53.
4. Ramani A, Dahoe AE. On the performance and mechanism of
brominated and halogen free flame retardants in formulations of
glass fiber reinforced poly(butylene terephthalate). Polym Degrad
Stab. 2014;104:71–86.
5. Balabanovich AI, Engelmann J. Fire retardant and charring effect
of poly(sulfonyldiphenylene phenylphosphonate) in poly(-
butylene terephthalate). Polym Degrad Stab. 2003;79:85–92.
123

Synthesis and performance of star-shaped aluminum phosphinate flame retardant
1409
˘
˘
6. Gallo E, Braun U, Schartel B, Russo P, Acierno D. Halogen-free
flame retarded poly(butylene terephthalate) (PBT) using metal
oxides/PBT nanocomposites in combination with aluminum
phosphinate. Polym Degrad Stab. 2009;94:1245–53.
24. Dogan M, Erdogan S, Bayraml E. Mechanical, thermal, and fire
retardant properties of poly(ethylene terephthalate) fiber con-
taining zinc phosphinate and organo-modified clay. J Therm Anal
Calorim. 2013;112:871–6.
7. Balabanovich AI. Thermal decomposition study of intumescent
additives: pentaerythritol phosphate and its blend with melamine
phosphate. Thermochim Acta. 2005;435:188–96.
8. Balabanovich AI. The effect of ammonium polyphosphate on the
combustion and thermal decomposition behavior of poly(-
butylene terephthalate). J Fire Sci. 2003;21:285–98.
25. Braun U, Schartel B. Flame retardancy mechanisms of aluminium
phosphinate in combination with melamine cyanuarte in glass-
fibre reinforced poly(1,4-butylene terephthalate). Macromol
Mater Eng. 2008;293:206–17.
26. Naik AD, Fontaine G, Samyn F, et al. Melamine integrated
metal phosphates as non-halogenated flame retardants: Syn-
ergism with aluminium phosphinate for flame retardancy in
glass fiber reinforced polyamide 66. Polym Degrad Stab. 2013;
98:2653–62.
27. Isitman NA, Kaynak C. Nanoclay and carbon nanotubes as potential
synergists of an organophosphorus flame-retardant in poly (methyl
methacrylate). Polym Degrad Stab. 2010;95:1523–32.
28. Isitman NA, Dogan M, Bayramli E, Kaynak C. The role of
nanoparticle geometry in flame retardancy of polylactide
nanocomposites containing aluminium phosphinate. Polym
Degrad Stab. 2012;97:1285–96.
¨
9. Hoerold S. Flame-retarding thermosetting compositions. US
Patent 6,420,459 (2002).
10. Stanley R, Springfield S. Polyester resins flame retarded by
poly(metal phosphinates)s. US Patent 4,180,495 (1979).
11. Brehme S, Koeppl T, Schartel B, et al. Competition in aluminium
phosphinate-based halogen-free flame retardancy of poly(-
butylene terephthalate) and its glass-fibre composites. E-Poly-
mers. 2014;14(3):729–47.
12. Braun U, Bahr H, Sturm H, Schartel B. Flame retardancy
mechanisms of metal phosphinates and metal phosphinates in
combination with melamine cyanurate in glass-fiber reinforced
poly(1,4-butylene terephthalate): the influence of metal cation.
Polym Adv Technol. 2008;19:680–92.
¨
29. Bauer H, Hoerold S, Krause W. Salts of alkyl esters of car-
boxyethyl(alkyl)phosphinic acid. US Patent 7,838,580 B2
(2010).
13. Liu XQ, Liu JY, Cai SJ. Comparative study of aluminum
diethylphosphinate and aluminum methylethylphosphinate-filled
epoxy flame-retardant composites. Polym Compos. 2012;33:918–26.
14. Kleiner HJ, Budzinsky W. Flame proofed polyester molding
composition. US Patent 6,013,707 (2000).
15. Kleiner H, Jenewein El, Wanzke, W. Alkyl-1-alkoxyethylphos-
phinous acid aluminium salts. EP 971,936 (2002).
16. Kleiner H. Aluminium salts of alkylhydroxymethylphosphinic
acids. US Patent 6,303,674 (2001).
17. Liu XQ, Liu JY, You QL, He D, Guo YH, Cakmak M. Influence of
structure of metal salts of phosphinates on the performance of flame-
retardant polymer. In: 30th International conference of the polymer
processing society (PPS-30). Cleveland, OH, USA; 2014.
30. Liu XQ, Liu JY, Chen J, Cai SJ, Hu CL. Novel flame-retardant
epoxy composites containing aluminium b-carboxylethyl-
methylphosphinate. Polym Eng Sci. 2015;55:657–63.
31. Liu JY, Chen J, Liu XQ. Synthesis of aluminum methylcyclo-
hexylphosphinate and its use as flame retardant for epoxy resin.
Fire Mater. 2014;38:155–65.
32. Wang Z, Lv P, Hu Y, Hu K. Thermal degradation study of
intumescent flame retardants by TG and FTIR: melamine phos-
phate and its mixture with pentaerythritol. J Anal Appl Pyrol.
2009;86:207–14.
33. Chen YH, Wang Q. Reaction of melamine phosphate with pen-
taerythritol and its products for flame retardation of polypropy-
lene. Polym Adv Technol. 2007;18:587–600.
¨
18. Koppl T, Brehme S, Wolff-Fabris F. Structure-property relationships
34. Lv P, Wang Z, Hu K, et al. Flammability and thermal degradation
of flame retarded polypropylene composites containing melamine
phosphate and pentaerythritol derivatives. Polym Degrad Stab.
2005;90:523–34.
35. Yang W, Lu H, Tai Q, Qiao Z, Hu Y, Song L, Yuen RK. Flame
retardancy mechanisms of poly(1,4-butylene terephthalate) con-
taining microencapsulated ammonium polyphosphate and mela-
mine cyanurate. Polym Adv Technol. 2011;22:2136–44.
36. Odinets IL, Antonov EA, Petrovsky PV, Morozik YuI, Krivolapov
YuA, et al. Interaction of 2-methyl-2,5-dioxo-1,2-oxaphospholane
with trimethylsilyl cyanide. Russ Chem Bull. 1994;43(1):121–4.
37. Williams D, Fleming I. Spectroscopic methods in organic
chemistry. New York: McGraw Hill; 2007.
38. Balabanovich AI. The effect of melamine on the combustion and
thermal decomposition behaviour of poly(butylene terephthalate).
Polym Degrad Stab. 2004;84:451–8.
39. Nagasaw Y, Ozaw HK. Fast thermolysis/FT-IR studies of fire-
retardant melamine-cyanurate and melamine-cyanurate contain-
ing polymer. J Anal Appl Pyrol. 1995;33:253–67.
of halogen-free flame-retarded poly(butylenes terephthalate) and
glass fiber reinforced PBT. J Appl Polym Sci. 2012;124:9–18.
19. Yang H, Song L, Tai Q, Wang X, Yu B, Hua Y, Yuen RK. Comparative
study on the flame retarded efficiency of melamine phosphate, mela-
mine phosphite and melamine. Polym Degrad Stab. 2014;105:248–56.
20. Brehme S, Schartel B, Goebbels J, Fischer O, Pospiech D, Bykov Y,
¨
Doring M. Phosphorus polyester versus aluminium phosphinate in
poly(butylene terephthalate) (PBT): flame retardancy performance
and mechanisms. Polym Degrad Stab. 2011;96:875–84.
¨
¨
21. Brehme S, Koppl T, Schartel B, Fischer O, Altstadt V, Pospiech
D, Doring M. Phosphorus polyester—an alternative to low-
¨
molecular-weight flame retardants in poly(butylene terephtha-
late)? Macromol Chem Phys. 2012;213:2386–97.
¨
22. Koppl T, Brehme S, Pospiech D. Influence of polymeric flame
retardants based on phosphorus-containing polyesters on mor-
phology and material characteristics of poly(butylene terephtha-
late). J Appl Polym Sci. 2013;128:3315–24.
23. Wang X, Song L, Yang H, Lu H, Hu Y. Synergistic effect of
graphene on antidripping and fire resistance of intumescent flame
retardant poly(butylene succinate) composites. Ind Eng Chem
Res. 2011;50:5376–83.
40. Sangeetha V, Kanagathara N, Sumathi R, Anbalagan G. Spectral
and thermal degradation of melamine cyanurate. J Mater.
2013;2013:1–7.
123