Aluminum-organophosphorus hybrid nanorods for simultaneously enhancing the flame retardancy and mechanical properties of epoxy resin

Article abstract of DOI:10.1039/c1jm13332b

New aluminum-organophosphorus hybrid nanorods (AOPH-NR) have been prepared by reacting aluminum hydroxide (ATH) with dibenzylphosphinic acid (DBPA) with aluminum hydroxide (ATH) and used to prepare nanocomposites with epoxy resin. In order to determine the structure-property relationship of these composites, several other phosphinic acids of the general formula (R(CH₂) _n)₂POOH (R = ester, allyl, nitrile, n = 1 or 2), and corresponding AOPHs were synthesized. FTIR, Raman, TGA, and XRD examinations showed that only AOPH-NR possesses a highly hybrid structure and high thermostability. SEM and TEM confirmed the nanorod morphology of AOPH-NR. The formation mechanism can be described as a decomposing-reforming process. This characteristic causes AOPH-NR to exhibit superior properties. Limiting oxygen index (LOI) determination and cone calorimeter analysis showed that the incorporation of only 4.25 wt% AOPH-NR remarkably improved the LOI value to as much as 28.0 and led to a 23% reduction in peak heat release rate (PHRR). Dynamic mechanical analysis (DMA) indicated that the mechanical properties of epoxy resin were also improved by incorporating AOPH-NR. In this way, the aluminum-organophosphorus hybridization via reacting ATH with specific organophosphinic acids shows promise as a means of improving flame retardancy and mechanical properties simultaneously. The thermal and anti-flaming properties of composites, combined with the properties of AOPHs, allowed us to discover the important role that the release and migration of phosphorus species plays in fire-retarding materials. This provides a new insight into the design of high-performance flame retardants.

Full text of DOI:10.1039/c1jm13332b

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links <
C
Journal of
Materials Chemistry
Cite this: J. Mater. Chem., 2012, 22, 2007
www.rsc.org/materials
PAPER
Aluminum–organophosphorus hybrid nanorods for simultaneously enhancing
the flame retardancy and mechanical properties of epoxy resin†
Ma Jiajun, Yang Junxiao,* Huang Yawen* and Cao Ke
Received 15th July 2011, Accepted 1st November 2011
DOI: 10.1039/c1jm13332b
New aluminum–organophosphorus hybrid nanorods (AOPH-NR) have been prepared by reacting
aluminum hydroxide (ATH) with dibenzylphosphinic acid (DBPA) with aluminum hydroxide (ATH)
and used to prepare nanocomposites with epoxy resin. In order to determine the structure–property
relationship of these composites, several other phosphinic acids of the general formula (R
(
CH
2
)
n
)
2
POOH (R ¼ ester, allyl, nitrile, n ¼ 1 or 2), and corresponding AOPHs were synthesized.
FTIR, Raman, TGA, and XRD examinations showed that only AOPH-NR possesses a highly hybrid
structure and high thermostability. SEM and TEM confirmed the nanorod morphology of AOPH-NR.
The formation mechanism can be described as a decomposing–reforming process. This characteristic
causes AOPH-NR to exhibit superior properties. Limiting oxygen index (LOI) determination and cone
calorimeter analysis showed that the incorporation of only 4.25 wt% AOPH-NR remarkably improved
the LOI value to as much as 28.0 and led to a 23% reduction in peak heat release rate (PHRR).
Dynamic mechanical analysis (DMA) indicated that the mechanical properties of epoxy resin were also
improved by incorporating AOPH-NR. In this way, the aluminum–organophosphorus hybridization
via reacting ATH with specific organophosphinic acids shows promise as a means of improving flame
retardancy and mechanical properties simultaneously. The thermal and anti-flaming properties of
composites, combined with the properties of AOPHs, allowed us to discover the important role that the
release and migration of phosphorus species plays in fire-retarding materials. This provides a new
insight into the design of high-performance flame retardants.
Introduction
compliance with various flame retardancy standards for poly-
20,21
mers.
The addition of fillers in such high amounts often leads
Natural and synthetic resins are rapidly taking the place of
traditional metallic and inorganic materials in various fields. This
is because of their excellent properties such as low density,
biocompatibility, biodegradability, clarity, and good weathering
behavior. However, during use, a large number of fires are
inevitable. This is because these resins are inherently flammable.
Enhancing the flame retardancy of these is becoming more and
more imperative, especially when they are used in fields within
the electronic industry and in building materials.
to a processing problem and marked deterioration in mechanical
properties of the composites. Second, the interfacial interaction
between hydroxides and organic polymers is poor. Even though
low filling levels can be achieved, the enhancement of flame
22–24
retardancy is still at the loss of mechanical strength.
Nano-ATH and cooperative use of ATH and other flame
retardants have been developed to solve these problems. Coop-
erative use of ATH and phosphorus-containing compounds,
such as triphenyl phosphate (TPP), can reduce the deleterious
effect that flame-retardant additives have on mechanical prop-
To date, a large number of flame retardants, including
10–13
and
1–5
6–9
minerals, phosphoric compounds, carbon materials,
22
erties. However, this technique can cause the bleeding-out of
25
phosphoric compounds on the surfaces of molded products. On
14–17
siliceous compounds,
have been used. Of these, aluminum
hydroxide (ATH) has received much attention because of its
environmental friendliness, low cost, low toxicity, low smoke,
the other hand, nano-ATH is difficult to disperse through the
polymer matrix due to its marked hydrophilicity and incom-
patibility between the surface characteristics of the substances.
18,19
and lack of corrosive gas;
however, it has two limitations.
First, more than 50 wt% of ATH is generally required for
26
This limits the enhancement in flame retardancy. There is still
a long way to go before the expected levels of performance are
reached.
School of Materials Science and Technology & State Key Laboratory
Cultivation Base for Nonmetal Composite and Functional Materials,
Southwest University of Science and Technology, Mianyang, 621010,
China. E-mail: yangjunxiao@swust.edu.cn; ywhuang1981@yahoo.cn.
Hybridization techniques, particularly inorganic–organic
hybridization, have brought a new perspective to the develop-
ment of high-performance flame retardants. Recent work has
focused on two layered clays, montmorillonite (MMT) and
† Electronic supplementary information (ESI) available: Fig. S1–S10. See
DOI: 10.1039/c1jm13332b
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 2007–2017 | 2007

View Article Online
2
7–31
layered double hydroxide (LDH).
Because of the relatively
cyanide, benzyl bromide, aluminum hydroxide, triethylamine,
and other general solvents were purchased from Chengdu
Kelong Reagent Co. Ammonium hypophosphite was purchased
from Hubei Hongjing Chemical Co., hexamethyldisilazane
(HMDS) from Hangzhou Guibao Chemical Co., allyl bromide
from Shandong Zouping Mingxing Chemical Co., methyl hex-
ahydrophthalic anhydride (MHHPA) from Puyang Huicheng
Chemical Co., and epoxy resin (diglycidyl ether of biphenol A,
DEGBA), with an epoxy value of 0.4083 mol/100 g were
purchased from Lanzhou Bluestar Resin Co.
small sizes of the layered clay particles, which are obtained by
intercalation, polymer-layered clay composites can markedly
improve the mechanical, thermal, and physical–chemical
properties of pure polymers. However, several studies on phos-
phorus-layered silicate systems have concluded that introducing
phosphorus species is not an efficient means of enhancing the fire
32–39
performance of epoxy-layered silicate nanocomposites.
As
a consequence, such hybrid systems have not been paid much
attention. The main reasons for the middling performance of the
phosphorus-layered silicates, in our opinion, are that general
hybridization leads to fair thermostability of phosphorus species
and that the layered structure may hamper the release and
migration of phosphorus species to the surface, decreasing the
efficiency of char formation.
Synthesis of dibenzylphosphinic acid (DBPA)
ꢀ
+
A mixture of 9.96 g (0.12 mol) H
ꢁ
2
PO
2
NH
4
and 34.2 g (0.21 mol)
HMDS was heated at 100–110 C in N
2
until the white ammonium
salt disappeared. Then 120 mL CH
2
Cl
2
and 14.4 mL (0.12 mol)
ꢁ
The pioneering work on MMT and LDH motivated us to
develop new aluminum–organophosphorous hybrid materials
benzyl bromide were added sequentially at 0 C, and the mixture
was stirred at room temperature for 12 h. Then, the reaction was
(
AOPHs) with marked thermostability and unique one-dimen-
ꢁ
cooled to 0 C, and another 34.2 g (0.21 mol) HMDS was injected
ꢁ
sional morphology. To the best of our knowledge, these types of
flame retardants have not yet been reported. AOPHs are superior
to other phosphorus-layered silicate systems in several ways. First,
the highly hybrid structure of AOPHs can be formed by covalent
bonding from nonhydrolytic and hydrolysis condensation reac-
immediately. Stirring was continued at 0 C for 2–3 h. Another
1
4.4 mL (0.12 mol) benzyl bromide was introduced and the
reaction was kept at room temperature for 12 h. Then the solution
was centrifuged and concentrated under reduced pressure. The
4
0,41
residue was dissolved in MeOH/CH Cl (1 : 1 v/v) for 30 min and
2 2
tions rather than ion-exchange,
This indicates greater ther-
the precipitated solid particles were filtered. The solution was
concentrated under reduced pressure. The residue was redissolved
mostability than the phosphorus-layered silicate system. As
a consequence, AOPHs release phosphorus compounds at rela-
tively high temperatures. Second, the hybridization of phosphorus
species and ATH allows for the formation of one-dimensional
ꢀ1
in CH
2
Cl
2
and washed three times with 4 mol L hydrochloric
. After
acid. The organic layer was dried over anhydrous MgSO
4
40
the removal of solvent under reduced pressure, the resultant solid
powder was purified by recrystallization from acetic acid and
materials. Suchone-dimensional structureswouldshowdifferent
flame-retarding behavior to layered silicates. In addition, like
modified layered silicates, this type of Al/P hybridization is able to
inhibit the bleeding out of phosphorus during processing. In this
way, it is superior to most combinations of ATH and TPP. Third,
the presence of organic moieties in the hybrid plays an important
role in the crystalline structure, morphology, dispersibility, and
thermal properties of AOPHs. AOPHs can be made to have a large
variety of compositions and a tunable number of hydroxyl groups,
as determined by changes in stoichiometry. This may be of
importance for tailoring the properties of AOPHs.
ꢁ
dried in a vacuum at 50 C, giving 24.3 g (82.5%) DBPA as a white
ꢁ
ꢁ
42
1
crystal; m.p.: 192 C (reported: 190 C ). H NMR (400 MHz,
DMSO-d , ppm) d: 7.19 (dd, J ¼ 17.4, 10.0 Hz, 6H), 7.12 (d, J ¼
6
1
2.5 Hz, 4H), 3.38 (d, 12H), 2.94 (s, 2H), 2.92 (d, 4H), 2.65–2.20
ꢀ
1
(
m, 3H). FTIR (neat, cm ): 2629, 2271, 1632, 1494, 1258, 1172,
+
54, 876, 786, 710. LC-MS (M-H ) m/z: 245.
9
Synthesis of bis(3-methoxy-3-oxopropyl)phosphinic acid
BMOPA)
(
In this work, we prepared a new aluminum–organophos-
phorus hybrid nanorod (AOPH-NR) by reacting aluminum
hydroxide with dibenzylphosphinic acid (DBPA). In addition,
a series of other phosphinic acids were produced with the same
The synthesis of BMOPA was almost the same as that of DBPA
except that the 14.4 mL benzyl bromide was replaced by 10.32 g
(
0.12 mol) methyl acrylate. After the reaction, the solution was
centrifuged and concentrated under reduced pressure. The
residue was dissolved in MeOH/CH Cl (1 : 1 v/v) for 30 min and
structural formula, of R POOH with DBPA and various func-
2
2
2
tional groups, such as ester, allyl, and nitrile groups. Corre-
sponding AOPHs were also prepared by reacting ATH with these
phosphinic acids. The chemical and layered structures and
thermal properties of the various flame retardants were investi-
gated. The flame retardants were blended with epoxy resin to
produce several composites. The morphology, thermal and
mechanical properties, flame retardancy, and structure–property
relationships of these composites were studied in detail.
the precipitated solid particles were filtered. The solution was
concentrated under reduced pressure. The residue was redis-
ꢀ1
2 2
solved in CH Cl and washed three times with 4 mol L
hydrochloric acid. The organic layer was dried over anhydrous
MgSO
4
. After the removal of solvent under reduced pressure, the
ꢁ
mixture was washed with hexane and dried in vacuum at 40 C.
BMOPA was produced as a white solid powder (17.99 g, 63.0%
ꢁ
1
6
yield); m.p.: 103 C. H NMR (400 MHz, DMSO-d , ppm) d:
ꢀ1
3
.66 (m, 2H), 2.87–2.18 (m, 2H). FTIR (neat, cm ): 2623, 2091,
+
743, 1440, 1252, 1180, 1046, 997, 900. LC-MS (M-H ) m/z: 237.
1
Experimental
43
Materials
Synthesis of diallylphosphinic acid (DAPA)
ꢀ
+
4
Unless otherwise noted, all reagents were purified using the
standard method before use. Methyl acrylate (MA), allyl
A solution of H
2
PO
2
NH
(4.98 g, 0.06 mol), ally bromide
(29.04 g, 0.24 mol), and HMDS (19.32 g, 0.12 mol) in 120 mL
2
008 | J. Mater. Chem., 2012, 22, 2007–2017
This journal is ª The Royal Society of Chemistry 2012

View Article Online
ꢁ
ꢁ
toluene was refluxed at 110 C under a flow of nitrogen for 12 h.
h, followed by a final curing at 140 C for 16 h, giving rise to solid
resins.
The solution was cooled to room temperature. After the removal
of the ammonium salts by centrifugation and concentration
under reduced pressure, the residue was redissolved in MeOH/
Instrumentation
CH
2 2
Cl (1 : 1 v/v) and the precipitated solid was filtered. After
concentration under reduced pressure, the resulting oil was dis-
Fourier transform infrared (FTIR) spectra were obtained on
a Perkin Elmer Spectrum One spectrophotometer with a resolu-
ꢀ1
solved in CH
hydrochloric acid. The organic layer was dried over anhydrous
MgSO . After the removal of solvent under reduced pressure, the
DAPA was obtained as a pale yellow oil (7.50 g, 85.6% yield). H
NMR (400 MHz, DMSO-d , ppm) d: 5.82–5.53 (m, 1H), 5.27–
2 2
Cl and washed three times with 4 mol L
ꢀ1
tion of 4 cm . Raman spectroscopic experiments were per-
4
formed on a inVia-Reflex confocal laser Raman spectrometer
ꢀ1
1
1
Renishaw Ltd) with a resolution of 2 cm . H NMR spectra
(
6
were collected on a Bruker Acance spectrometer (400 MHz)
using tetramethylsilane (TMS) as an internal reference and
ꢀ1
4
1
.86 (m, 2H), 2.81–1.89 (m, 2H). FTIR (neat, cm ): 2622, 2276,
+
638, 1230, 1160, 1061, 971, 918. LC-MS (M-H ) m/z: 145.
6
deuterated dimethyl sulfoxide (DMSO-d ) as a solvent. Mass
spectra (m/z) were recorded on a Varian 1200LC/MS. Powder
X-ray diffractometer (XRD) patterns were recorded on a Pan-
alytical X’Pert Pro diffractometer with Cu-Ka radiation.
Dynamical mechanical analysis (DMA) measurements were
performed on a TA Q800 dynamical mechanical analyzer in
three-point bending mode at a frequency of 1 Hz. Thermogra-
vimetric analysis (TGA) was performed on a TA SDT Q600
Synthesis of bis(2-cyanoethyl)phosphinic acid (BCEPA)
The synthesis of BCEPA was also the same as that of DBPA
except that 8 mL (0.11 mol) allyl cyanide was used instead of 14.4
mL benzyl bromide. The solution was centrifuged and concen-
trated under reduced pressure. The residue was dissolved in
2 2
MeOH/CH Cl (1 : 1 v/v) for 30 min and the precipitated solid
ꢁ
ꢀ1
simultaneous DTA-TGA at a heating rate of 10 C min for
ꢀ1
particles were filtered. After concentration under reduced pres-
sure, the residue was redissolved in CH Cl and washed three
ꢁ
2
AOPHs and 20 C min for composites in an N and air
2
2
atmosphere. Limiting oxygen index (LOI) were measured on
a LFY-606 oxygen index instrument (Qingdao, China) with a test
specimen bar of 120 ꢂ 6 ꢂ 3 mm according to ASTM D-2863.
Scanning electron microscopy of AOPHs was recorded on an
S-4800 (Hitachi Ltd) field emission scanning electron micro-
scope. Scanning electron microscopy of cross sections of
composites were recorded on a TM-1000 (Hitachi Ltd) scanning
electron microscope at an accelerating voltage of 10 kV.
Morphological observation for AOPHs and their dispersion in
the epoxy resin was performed with Tecnai G2 F20 S-TWIN
ꢀ1
times with 4 mol L hydrochloric acid. The organic layer was
dried over anhydrous MgSO . After the removal of solvent under
4
reduced pressure, washing with hexane and drying in vacuum at
ꢁ
4
0 C, BCEPA was produced as a white solid powder (17.6 g,
1
ꢁ
8
6
5.1% yield); m.p.: 100 C. H NMR (400 MHz, DMSO-d , ppm)
d 2.69–2.54 (m, 1H), 2.47 (d, J ¼ 15.6 Hz, 1H), 2.06–1.76 (m, 1H).
ꢀ1
FTIR (neat, cm ): 2648, 2248, 1260, 1177, 1093, 1017, 982, 897.
+
LC-MS (M-H ) m/z: 171.
Preparation of AOPHs
(
FEI Company) transmission electron microscopy. The epoxy
composite samples were microtomed with a diamond knife at
room temperature into sections with a nominal thickness of 100
nm. The flammability of the epoxy resins and their composites
were characterized using a cone calorimeter performed in Stan-
ton Redcroft, UK device according to ISO 5660 with an incident
To prepare DBPA modified ATH, 2.8 g (19.2 mmol) DBPA was
3
added to a suspension of 1.0 g Al(OH) (a commercial product
ꢁ
with a size of about 10 mm, dried at 120 C for 12 h prior to use)
in 30 mL of acetic acid. The resultant suspension was stirred at 80
ꢁ
2
C in N for more than 24 h until the appearance of white floating
ꢀ
2
flux of 35 kW m using a cone-shaped heater. Results were
solids. After the reaction, the white powder was collected by
filtration, washed with water, ethanol, and alcohol, in that order,
found to be reproducible to within ꢃ10%.
ꢁ
and then dried at 50 C for 12 h in vacuum.
DAPA, BMOPA, and BCEPA modified ATH were prepared
in the same way. One gram of ATH was reacted with a solution
of phosphinic acid (19.2 mmol phosphinic acid was dissolved in
Results and discussion
Synthesis of phosphinic acids
3
0 mL of acetic acid) at room temperature in N for 36 h. After
2
The use of phosphoric acids to prepare organically modified
aluminum phosphates or organophosphorus modified aluminas
has been the subject of intense research. Florjanczyk et al. have
prepared aluminum phosphates from the reaction of diphenyl-
the reaction, the white powder was collected by filtration, washed
with deionized water, washed sequentially with ethanol and
ꢁ
acetone, and then dried at 50 C for 12 h in vacuum.
3 3 3
phosphoric acid (DPPA) with AlMe , AlEt , Al(tBu) and
Preparation of epoxy resin/AOPHs composites
40
boehmite. Vioux et al. have reported the preparation of
To prepare epoxy resin/AOPHs composites, AOPHs were
dispersed in toluene by sonication. Epoxy resin in 50 mL of
toluene was mixed with the APOHs solution, followed by stirring
under vacuum at room temperature for 30 min. After concen-
tration by rotary evaporation, the curing agent methyl hexahy-
drophthalic anhydride (MHHPA) and the catalyst triethylamine
organically modified alumina by grafting with phenylphosphonic
44
acid and its organic-soluble ester derivatives. Hix et al. have
studied the hydrothermal reaction of ATH with mixtures of
phosphorous and methylphosphonic acids, giving microporous
41
aluminum methylphosphonate. In our case, phosphinic acids
were used instead of phosphoric acids, in terms of the relatively
high thermal stability of the P–C bond relative to that of the
P–O–C bond. To the best of our knowledge, no such structure
(
TEA) were added to epoxy resin/AOPHs system sequentially.
ꢁ
The reaction mixtures were pre-cured in a mould at 110 C for 1
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 2007–2017 | 2009

View Article Online
has been reported previously in aluminum–organophosphorus
hybrids.
hydroxide (ATH) via nonhydrolytic and hydrolysis condensation
48,49
accompanied with forming thermally stable Al–O–P bonds.
In the present work, DBPA was synthesized by a one-pot
intermolecular Arbusov-type reaction, as reported by Bujard and
The reaction of phosphinic acid with ATH was carried out via
treating a suspension of ATH with the phosphinic acids in
appropriate organic mediums (acetic acid or other organic acids).
The chemical structure of AOPH-NR was characterized by
FTIR (Fig. 1). The appearance of a characteristic absorption
band corresponding to phenyl groups was observed in Fig. 1a,
preliminarily supporting the reaction of ATH with DBPA. By
carefully comparing the FTIR spectra of ATH, DBPA, and
43,45
Boyd,
between bis(trimethylsilyloxy) phosphine (BTSP) and
benzyl bromide (Scheme 1). The intermediate, BTSP, was
produced in situ from the reaction of ammonium phosphinate
and HMDS. This method had the advantage of synthesizing
a variety of compounds containing phosphorus–carbon bonds,
involving symmetrical and asymmetrical phosphonic and phos-
phinic acids. Other phosphinic acids were prepared using
a similar method. BTSP was reacted with alkyl or allyl halides, a,
b-unsaturated ketones, and other unsaturated compounds
ꢀ
1
AOPH-NR in the range of 950–1200 cm (Fig. 1b), it can be seen
that the absorption bands of P]O stretching at nearly 1200
ꢀ1
ꢀ1
cm , P–OH stretching at 950–1000 cm , and Al–OH stretching
ꢀ1
(
Scheme 1). The structures of all the phosphinic acids were
at 1061 cm greatly decreased or nearly disappeared. These
results suggest the formation of P–O–Al bonds via a condensa-
tion reaction between ATH and phosphinic acids. The formation
of P–O bonds is further supported by the presence of their
1
confirmed by FTIR, H NMR, and LC-MS (Fig. S1–S3, ESI†).
The characteristic absorption bands of C]C, C]O, benzene
rings, and ChN were observed in FTIR spectra of the four
phosphinic acids. Signals indicating the hydrogens on –CH
2
– and
characteristic absorption bands in the range of 1000–1100
ꢀ1 44,50,51
1
– were found in the H NMR spectra of all phosphinic
–
CH
2
CH
2
cm .
A great decrease in the band generated by P]O
stretching and the disappearance of the band generated by Al–O
stretching was observed for AOPH-NR, suggesting the existence
acids, strongly suggesting the generation of P–C bonds. In
addition, the molecular weights found in MS were identical to the
calculated molecular weights.
of a large number of bidentate PO or chelating monodentate
2
40,52
The reaction of aluminum hydroxide (ATH) with DBPA
produced AOPH-NR. AOPHs prepared from other phosphinic
acids were designated as AOPH-C1, AOPH-C2, and AOPH-C3
as shown in Scheme 1. As expected, AOPH-NR possessed
a unique structure and properties, which allowed us to undertake
a careful investigation of the structure–property relationship,
detailed in the following sections.
structures.
Similar results were also found in the spectra of
other AOPHs (Fig. 1c–e). Nevertheless, a relatively small
decrease in P]O was observed for other AOPHs, which is
indicative mainly of the pendant monodentate structure. Further
indications of hybrid structure were provided by Raman spec-
troscopy (Fig. 2). The characteristic bands of organic groups for
AOPH-NR were observed. Several new bands in the range of
ꢀ
1
1
000–1400 cm , which were not present in the ATH spectrum,
were seen. These bands probably correspond to the vibrations of
the PO unit. In addition, the near-disappearance of the vibra-
Characterization of AOPH-NR
2
ꢀ1
tional modes of the P]O bond at 1280–1295 cm and the
Typical ATH crystalline is composed of aluminum octahedral
32,46,47
appearance of asymmetric and symmetric P–O vibration bands
53,54
layers with hydroxyl groups on either side.
The lack of
ꢀ
1
ꢀ1
+
of O–P–O at 819 cm and 763 cm were observed.
In
ꢀ
exchangeable cations such as Na makes it difficult to perform an
ion exchange process using cetrimonium bromide (CTAB) or any
other cation modifier like MMT and LDH. The use of aqueous
solutions of phosphonic acid, nevertheless, may result in the
reaction of organophosphorus compounds with aluminum
1
particular, intense P–O bands were observed at 1000–1400 cm ,
55
possibly from the formation of Al–O–P–O–Al. In agreement
with the analysis of FTIR, the above results are indicative of
a mainly bidentate PO
2
or chelating monodentate structure for
AOPH-NR. For other AOPHs, the weak P–O bands and
remaining P]O bands indicate
structure.
a pendant monodentate
The thermal properties of ATH and AOPHs was evaluated by
thermogravimetric analysis (TGA) and derivative thermogravi-
metric analysis (DTG) under a nitrogen atmosphere (Fig. 3 and
S4, ESI†). The decomposition of ATH was a single-step reaction.
This can be attributed to the loss of external surface water,
gallery water, and the partial dehydration reaction of ATH. In
contrast, AOPH-NR decomposition required two steps. The
second decomposition step mainly involved the degradation and
volatilization of organophosphorus molecules, as evidenced by
the complete decomposition of DBPA, shown in its TGA
(
Fig. S5, ESI†). AOPH-NR decomposed, releasing its organo-
phosphorus species, at a higher temperature than DBPA. Other
AOPHs also showed higher temperature of releasing their
organophosphorus species relative to the corresponding phos-
phinic acids. These indicate that hybridization effectively
improved the thermal stability and releasing temperature of the
organophosphorus species. The decomposition temperature and
Scheme 1 Schematic synthetic routes to four phosphinic acids and
corresponding AOPHs.
2
010 | J. Mater. Chem., 2012, 22, 2007–2017
This journal is ª The Royal Society of Chemistry 2012

View Article Online
Fig. 1 Chemical structure as determined by FTIR: (a) FTIR spectra of ATH, AOPH-NR, and other AOPHs throughout the whole range. (b–e)
ꢀ
1
Detailed analysis of AOPH-NR, other AOPHs, phosphinic acids, and ATH within the range of 950–1200 cm . (f) Schematic illustration of the possible
bonding configurations of the P–O–Al bond.
bridging bidentates in AOPH-NR, which have relatively good
thermal stability relative to mondentate bridging structures.
ATH crystals have three main polymorphs, gibbsite, bayerite,
56
and nordstrandate. Because of the strong hydrogen bonds
between hydroxide layers, these layers are bonded tightly
together with a very narrow basal spacing of 0.485 nm. The
change in crystal structure, a consequence of the chemical
incorporation of phosphinic acids into ATH, was examined by
powder X-ray diffraction (XRD). Fig. 4 shows the XRD patterns
of ATH, AOPH-NR and other AOPHs. The near-disappearance
of the XRD pattern of ATH in the AOPH-NR images indicates
a significant change in crystal structure. One possible explanation
is that the high degree of hybridization taking place in AOPH-
NR, as shown in FTIR and Raman, has destroyed the inherent
crystal structure of ATH. Other AOPHs have the same peaks as
ATH at 2q ¼ 18.3ꢁ and 20.6 , suggesting that they share the
inherent crystal lattice structure of ATH (Fig. S6, ESI†). This
strongly indicates the organophosphorus modified aluminum
structure of other AOPHs. Another notable observation made of
Fig. 2 Raman spectra of ATH, AOPH-NR, and other AOPHs.
ꢁ
ꢁ
other AOPHs is that new peaks appeared at angles below 10 .
Calculations made using the Bragg equation, d ¼ nl/(2Sinq),
indicates that they had wider basal spacing. This indicates an
32
intercalated structure. This conclusion was further supported
by TEM imaging, which showed the tactoids of a few platelets,
characteristic of intercalated structures (Fig. S7, ESI†).
Fig. 3 TGA curves of ATH, AOPH-NR, and other AOPHs in nitrogen.
residual weight of AOPH-NR showed a certain degree of
decrease relative to ATH. This was because of the incorporation
of organic moieties. Unlike other AOPHs, AOPH-NR showed
little weight loss during the first decomposition step, indicating
that the greatest number of hydroxyl groups had been consumed.
As noted in FTIR, this implies a bidentate PO and chelating
2
monodentate structure. In addition, AOPH-NR showed a higher
T
d
in the second decomposition than other AOPHs. This result is
associated with the higher T of DBPA. This can also be ascribed
d
to the presence of a large number of chelating mondentates and
Fig. 4 XRD patterns of ATH, AOPH-NR, and other AOPHs.
J. Mater. Chem., 2012, 22, 2007–2017 | 2011
This journal is ª The Royal Society of Chemistry 2012

View Article Online
The destruction of the ATH layered crystalline structure in
AOPH-NR means that the intercalated structure hardly forms.
ꢁ
Unlike other AOPHs, a strong reflection at 2q ¼ 6.27 (a ¼ 14.8
ꢀ
A) appears in the XRD pattern of AOPH-NR, implying a kind of
layered or fibrous morphology. In agreement with this idea, SEM
and TEM images (Fig. 5) clearly show the AOPH-NR to have
a nanorod morphology, with a rod diameter of <100 nm. Such
a regular nanorod morphology has not been observed in previous
works related to the organophosphorus-modified aluminum. In
addition, this kind of morphology was not observed for ATH or
for other AOPHs. These exist as solid agglomerates of relatively
large size (Fig. S6 and S7, ESI†). The preliminary explanation for
the formation of nanorods is that the destruction of the ATH
crystal structure gives rise to polymeric chains with Al–O–P
bridging structures, which form a close-packed columnar struc-
ture. According to this mechanism, octahedrally coordinated Al
atoms bridged by three dibenzylphosphate (DBP) ligands could
Fig. 6 Structure and mechanism of formation of AOPH-NR and other
AOPHs.
EP/AOPHs are only microcomposites; large particles and even
agglomerates were observed in the SEM and TEM images
40
be formed. Absorption bands corresponding to such ligands
were actually observed in FTIR. The parameter a (1.480 nm) for
AOPH-NR is in agreement with the diameter of the polymeric
(
Fig. S8–S9, ESI†). In addition, the neat epoxy resin and
EP/ATH composite displays a relatively smooth fracture surface,
although some plastic-deformed veins corresponding to the
shrinkage deformation of the film did appear (Fig. S8, ESI†).
With the addition of AOPH-NR, irregular plastics deformation
morphology, void growth, debonding of the particles from the
matrix, and highly rough fracture surfaces with a platelet-like
morphology were observed. These phenomena strongly suggest
that the toughening mechanism was dominant in the polymer–
40
3
chains with Al(DBP) repeating units. This provides further
evidence for the proposed structure. In addition, two weak
ꢁ
reflections at 2q ¼ 22.0–25.0 with a d-spacing distance of 0.389
nm and 0.371 nm were observed (Fig. 4), which implies the
57–60
presence of p–p stacking interactions.
When the role of p–p
stacking interactions in directing supramolecular architectures is
taken into account, it can be supposed that the assembly of
polymeric chains could be driven by p–p stacking interac-
64
particle composites. They are also indicative of improved
mechanical properties. In comparison, most of the other
EP/AOPH composites showed smooth fracture surfaces, indi-
cating limited improvement in mechanical properties.
61–63
tions.
Nevertheless, the exact formation mechanism of such
a rod-like morphology still requires further study, in which our
group is currently engaged. In conclusion, all of the above
findings involving FTIR, Raman, TGA, XRD, SEM, and TEM,
have allowed us to describe the structure of AOPH and the
method by which that structure forms (Fig. 6).
Curing behavior
The curing behavior of the EP composites is examined by DSC
(
Fig. S10, ESI†). Large amounts of heat were released from the
Morphology of EP/AOPH nanorod nanocomposite
ring-opening reaction, more by EP/AOPH-NR than by other
EP/AOPHs. This indicates a high degree of curing. In compar-
ison, the small amount of heat released by other EP/AOPH
composites is indicative of a relatively low degree of curing. In
addition, the difference between the heat releasing temperature
of the EP/AOPH-C2 composite and that of the EP/AOPH-C1
and EP/AOPH-C3 composites was observed. This can be
Scanning electron micrography (SEM) and transmission electron
microscopy (TEM) were used to determine the microscopic
morphologies of neat EP and EP/AOPHs composites. As shown
in Fig. 7b–c, both SEM and TEM images of the EP/AOPH-NR
composite show a good dispersion of AOPH-NR in the
epoxy resin, indicating the formation of a nanocomposite. Other
Fig. 5 Microscopic morphology of AOPH-NR: (a) SEM image and (b) TEM image.
012 | J. Mater. Chem., 2012, 22, 2007–2017 This journal is ª The Royal Society of Chemistry 2012
2

View Article Online
Fig. 7 SEM images of cross section of (a) neat EP and (b) EP-4.25 wt% AOPH-NR composite. (c) TEM image of EP-4.25 wt% AOPH-NR composite.
ascribed to the reaction between allyl groups and epoxy resin.
The C]C bonds in AOPH-C2 could polymerize by themselves
and, at the same time, interfere with the crosslinking of epoxy by
reacting with the epoxy groups. The improved dispersion of
AOPH-C2 in the epoxy resin could also be a result of this
groups, dispersibility, and interfacial interactions. The incorpo-
ration of ATH and other AOPHs into EP, regardless of their
dispersibility in EP and the flexibility of organic groups in
AOPHs cannot efficiently increase the glass transition. This trend
allowed us to suggest that the interfacial interaction between
matrix and fillers, which was only a physical interaction, is not
strong enough to increase T . The obvious decrease of T for
65
reaction.
g
g
EP/ATH composites was attributed to the large size of ATH or
inherent aggregates (Fig. S8 and S9, ESI†), leading to a decrease
in the number of particles in the polymer and to an increased
confinement effect between particles. It also caused an increase in
the number of defects, amount of free volume, and the mobility
Dynamic mechanical analysis, DMA
Dynamic mechanical analysis (DMA) is an effective tool for the
characterization of the interfaces of filled polymer systems,
especially the damping spectra. Fig. 8a shows the tan (d) of neat
epoxy resin, EP/ATH, EP/AOPH-NR, and other EP/AOPHs
composites with filler loading of 4.25 wt% as a function of
g
of the chains. The T of the epoxy/AOPH-C2 composite mark-
edly decreased. As described above, this can be ascribed mainly
to the lowering of the crosslink density of the epoxy composite.
temperature. The glass transition temperatures (T
g
) of neat EP
and EP composites were extracted in terms of the peak temper-
Further consideration is given by the T dispersion depicted in
g
66,67
ature of tand ꢀ T curves.
Fig. 8a, which has been found to be correlated to the interfacial
structure. The addition of AOPH-NR broadens the glass tran-
sition and decreases its intensity with respect to the neat epoxy
resin. In addition, a shoulder of the principal peak appears in the
The incorporation of particles can cause the T
g
of polymers to
increase, decrease, or remain constant, depending on the
dispersion of particles in the matrix and whether the polymer–
68,69
ꢁ
particle interactions are strong, weak, or repulsive.
The
range of 60–100 C. This broadening of relaxation times indi-
flexibility of the organic groups grafted onto the fillers has an
impact on the modulus as well when several organic groups are
cated that the AOPH-NR, which are compatible with the epoxy
resin and are small in size, are more readily incorporated into the
network structures than other fillers. This causes increased
heterogeneity within the network structure.
present. The EP/AOPH-NR nanocomposite has a T close to that
g
of neat epoxy resin but higher than that of the EP/ATH
composite (Fig. 8). This indicates that the addition of AOPH-NR
causes a greater confinement effect between chains than ATH
does. This comes from the net effect of the flexibility of organic
The results obtained from tan (d) curves allowed us to consider
the affecting factors of modulus. From Fig. 8b, the incorporation
of ATH and AOPH-NR in the epoxy resin leads to an increase in
Fig. 8 Typical DMA results of neat epoxy resin, EP/AOPH-NR nanocomposite and other EP composites studied: loss factor curve (a) and storage
modulus curves (b).
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 2007–2017 | 2013

View Article Online
Fig. 9 TGA thermograms of neat epoxy resin and epoxy composite with a loading of 4.25 wt% AOPH-NR and other AOPHs in (a) nitrogen and (b) air
ꢁ
ꢀ1
at a heating rate of 20 C min
.
ꢁ
storage modulus at 40 C (the glassy state). In the glassy state,
the modulus is mainly dependent on the stiffness of filler and the
interactions between the matrix and the filler. The increase can be
attributed to the greater stiffness of ATH and AOPHs relative to
the epoxy resin because the interactions between the matrixes
and fillers were here found to be less strong. The EP/AOPH-NR
nanocomposite was found to have a considerably higher storage
modulus in contrast to the EP/ATH composite, thus supporting
that the hybridization of organophosphorus with ATH is able to
enhance the reinforcement effect. This is in good agreement with
the evidence derived from the SEM images of the EP/AOPH
composites. In addition, the EP/AOPH-NR nanocomposites
show a higher storage modulus relative to other EP/AOPHs
composites. This can be partly ascribed to good dispersibility (as
composites has never been addressed. However, it has been
shown that selected synthetic materials can efficiently improve
71–73
the thermal stability of epoxy resin.
The thermal stability of neat EP, EP/AOPH-NR nano-
composites, and other EP/AOPH composites is given in Fig. 9
and Table 2. Neat epoxy resin, which was cured by hexahy-
ꢁ
drophthalic anhydride, begins its weight loss at 402.4 C in N
ꢁ
2
and at 373.5 C in air. It shows a single-step weight loss in N
2
and
a primarily two-step weight loss in air, in which the second
decomposition step involves the skeletal degradation of cured
22
epoxy resin. A similar trend was observed for the EP/AOPH-
NR nanocomposite. From TGA, the addition of AOPH-NR was
found to significantly facilitate this decomposition in both N
2
and air. It also leads to an increase in residual weight. The
facilitation of degradation of AOPH-NR was correlated to
the release of organophosphorus species as described above. The
migration behavior of the phosphorus species, in this case, may
be as significant as the releasing behavior. Both good dispersion
and nano-scale particle size are beneficial to the migration of the
phosphorus species. Other AOPHs at 4.25 wt% additive loading
evidenced by SEM and T dispersion) and a large number of rigid
g
phenyl groups in AOPH-NR. Most importantly, the nanorod
morphology of AOPH-NR may play an important role owing to
70
the large surface area and aspect ratios. The relatively high
modulus of the EP/AOPH-C2 composite could be attributed to
an improved interfacial interaction and dispersion caused by the
reaction between the allyl groups and epoxy resin. Nevertheless,
this reaction induces incomplete crosslinking in the epoxy resin,
leading to a pronounced decrease in the modulus with increasing
the temperature (Fig. 8b).
2
lead to a slight decrease in decomposition temperature in N .
Some AOPHs even caused a decrease in char residue. In addi-
tion, the incorporation of other AOPHs caused an increase in
decomposition temperature in air. The residual weight in air was
considerably lower than in N2.
The flame retardancy of EP/AOPH composites was evaluated
using a limiting oxygen index (LOI) and cone calorimeter. As
shown in Table 2, the loading of 4.25 wt% ATH only slightly
Thermal properties and flame retardancy
Many papers have been written on the thermal degradation of
16,17
epoxy resin and its corresponding composites.
However, the
2 3
improved the LOI value, to 23.6. Although the water and Al O
role of aluminum–organophosphorus hybridization on the
thermal stability or flame retardancy of the corresponding
simultaneously produced from the decomposition of ATH can
slow down the combustion of the matrix, they failed to stop it
completely in this case. In order to achieve the flame retardancy
74,75
standards for epoxy resin, at least 50 wt% of ATH is required.
Table 1 TGA data for ATH, AOPH-NR, and other AOPHs
The incorporation of only 4.25 wt% AOPH-NR improved the
LOI value to as much as 28 (an increase ratio of 26.1% than neat
EP), showing self-extinguishing behavior. Besides, the addition
of single phosphorus compounds gives a lower increase ratio of
First stage
Second stage
ꢁ
ꢁ
Sample
T
d
/ C
Mass loss/%
T
d
/ C
Mass loss/%
76,77
ATH
290.0 ꢃ 2
268.2 ꢃ 2
280.7 ꢃ 2
278.0 ꢃ 2
280.0 ꢃ 2
27.7 ꢃ 0.3
4.5 ꢃ 0.3
—
—
the LOI value of about 14.0–14.4%.
Comparison of ATH,
AOPH-NR
AOPH-C1
AOPH-C2
AOPH-C3
495.9 ꢃ 2
382.0 ꢃ 2
488.6 ꢃ 2
444.2 ꢃ 2
65.5 ꢃ 0.3
30.6 ꢃ 0.3
26.8 ꢃ 0.3
14.8 ꢃ 0.3
single phosphorus compounds, and AOPH-NR indicated an
improvement in flame retardancy due to the hybridization of
phosphorus compounds. The addition of other AOPHs showed
a slight increase of LOI value. The heat release curves for epoxy
16.3 ꢃ 0.3
19.5 ꢃ 0.3
22.9 ꢃ 0.3
2
014 | J. Mater. Chem., 2012, 22, 2007–2017
This journal is ª The Royal Society of Chemistry 2012

View Article Online
Table 2 Summary of TGA data for neat epoxy resin, EP/AOPH-NR nanocomposite and other EP composites
In N
2
condition
In air condition
Char residue at
ꢁ
Char residue at
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
Sample
T
0
/ C
T
max/ C
500 C/%
T
0
/ C
T
max/ C
500 C/%
LOI/%
Neat EP
EP-4.25% ATH
EP-4.25% AOPH-NR
EP-4.25% AOPH-C1
EP-4.25% AOPH-C2
EP-4.25% AOPH-C3
402.4 ꢃ 2
395.9 ꢃ 2
377.6 ꢃ 2
400.1 ꢃ 2
397.7 ꢃ 2
398.5 ꢃ 2
423.9 ꢃ 2
415.8 ꢃ 2
403.7 ꢃ 2
422.5 ꢃ 2
420.4 ꢃ 2
423.6 ꢃ 2
5.6 ꢃ 0.3
5.2 ꢃ 0.3
12.2 ꢃ 0.3
4.7 ꢃ 0.3
15.7 ꢃ 0.3
8.9 ꢃ 0.3
373.5 ꢃ 2
380.6 ꢃ 2
349.8 ꢃ 2
390.5 ꢃ 2
373.9 ꢃ 2
384.9 ꢃ 2
422.3 ꢃ 2
420.5 ꢃ 2
394.8 ꢃ 2
424.2 ꢃ 2
421.3 ꢃ 2
424.1 ꢃ 2
19.8 ꢃ 0.3
18.4 ꢃ 0.3
22.1 ꢃ 0.3
20.2 ꢃ 0.3
24.3 ꢃ 0.3
19.9 ꢃ 0.3
22.2
23.6
28.0
23.4
23.6
23.8
80,81
organophosphorus hybrid structure.
First of all, these features
improve the dispersibility of AOPH-NR, which is probably linked
to high fire performance. Second, in otherwise identical blends,
the aluminum–organophosphorus hybridization would improve
the flame retardancy arising from their counteraction of the
promotion of char formation, as evidenced by the increased char
residue of the EP/AOPH-NR nanocomposite relative to the EP/
ATH composite. This, in addition to the observed facilitation of
degradation in TGA, underlined a proposed mechanism of flame
retardancy based on char formation through the cleavage of P–O–
C and P–C bonds, followed by dehydration processes promoting
char formation. The phosphorus species hence provided superior
fire resistance by lowering decomposition temperatures of epoxy
resin and increasing char yields. The nanorod morphology of the
as-prepared aluminum–phosphorus hybrids may enable conve-
nient release and migration of phosphorus species. This favors the
improvement of the composite’s fire-retarding properties. It was
also found that the fire properties of the phosphorous-containing
samples generally deteriorated in the presence of clay, typically
MMT, even when the phosphorus compounds were intercalated
Fig. 10 Heat release rate curve for neat epoxy resin and composites with
a loading of 4.25 wt% ATH, AOPH-NR, and other AOPHs.
resin with AOPH-NR and other AOPHs at a loading of 4.25 wt%
are given in Fig. 10 and a summary of the cone data is given in
Table 3. As shown in Fig. 10, EP/AOPH-NR nanocomposite
shows a pronounced reduction (approximately 23%) in peak heat
release rate (PHRR), as indicated in Table 3, but no reduction is
noted for other EP/AOPHs composites. This reduction is not very
82
into the clay. MMT showed advantages in inhibiting the release
of heat because of its layered structure. Nevertheless, this struc-
ture, we suspect, hampers the migration of phosphorus species to
the surface. The DBPA used to prepare AOPH-NR was not
considered truly flame retardant because of its low decomposition
temperature. However, when hybridized with ATH, its phos-
phorus-releasing temperature was increased. This may also have
contributed to the good flame retarding performance observed.
29,78,79
large compared to what has been seen with MMT.
As shown
in Table 3, the tign jumps quite a bit, from 48 to 79 and the time to
PHRR is prolonged by 225 when AOPH-NR is added to EP. This
delay in ignition is crucial to saving lives and property. Other
parameters, such as average mass loss rate (AMLR) showed
a reduction (more than 50%) relative to EP. It is worth noting that
this reduction is much more pronounced than that of PHRR. For
nanocomposites, the change in PHRR generally reflects a change
in the mass loss rate. However, that is clearly not the case here,
possibly implying a different flame retarding mechanism.
Conclusions
A new aluminum–organophosphorus hybrid nanorod was
prepared by reacting ATH with dibenzylphosphinic acid
(DBPA). FTIR and Raman measurements suggest that the
The flame retarding efficiency of AOPH-NR, we suggest, can
be ascribed to its nanorod morphology and to its aluminum–
a
Table 3 Cone calorimeter data for neat epoxy resin and composites with a loading of 4.25 wt% ATH and AOPHs
ꢀ
2
ꢀ2
2
ASEA/m kg
ꢀ1
ꢀ1
Sample
t
ign/s
t
PHRR/s
PHRR/kW m
THR/MJ m
AMLR/g s
Neat EP
48
79
80
58
78
175
225
185
160
195
1023
789
1092
1063
1024
109
101
107
99
343
761
608
314
265
0.2618
0.1102
0.1635
0.1313
0.2858
EP-4.25% AOPH-NR
EP-4.25% AOPH-C1
EP-4.25% AOPH-C2
EP-4.25% AOPH-C3
106
a
tign: time to ignition; tPHRR: time to PHRR; PHRR: peak heat release rate; THR: total heat release; ASEA: average specific extinction area; AMLR:
average mass loss rate.
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 2007–2017 | 2015

View Article Online
organophosphorus species were incorporated into ATH via the
formation of Al–O–P bonds, mainly producing the chelating
mondentate and bridging bidentate structure. These corre-
sponded to the destruction of the ATH crystalline structure, as
evidenced by XRD. The TGA of DBPA and AOPH-NR indi-
cated that AOPH-NR has a favorable phosphorus releasing rate
and a high phosphorus releasing temperature for multiple
phosphorus species. SEM and TEM images of AOPH-NR show
that it has nanorod morphology while other AOPHs displayed
a solid agglomerates morphology. XRD examination prelimi-
narily suggests a decomposing–reforming mechanism that
reacting phosphinic acids with ATH resulted in the decomposi-
tion of ATH crystal structure into polymeric chains with Al
6 P. Song, Y. Shen, B. Du, M. Peng, L. Shen and Z. Fang, ACS Appl.
Mater. Interfaces, 2009, 1, 452–459.
7
8
G. Jinping and C. Guoqiang, Fire Mater., 2010, 34, 261–270.
H. Lim and J. Y. Chang, J. Mater. Chem., 2010, 20, 749–754.
9 J. Kim, S. Yoo, J. Y. Bae, H. C. Yun, J. Hwang and B. S. Kong,
Polym. Degrad. Stab., 2003, 81, 207–213.
1
1
1
1
0 A. L. Higginbotham, J. R. Lomeda, A. B. Morgan and J. M. Tour,
ACS Appl. Mater. Interfaces, 2009, 1, 2256–2261.
1 S. Nie, L. Song, Y. Guo, K. Wu, W. Xing, H. Lu and Y. Hu, Ind. Eng.
Chem. Res., 2009, 48, 10751–10758.
2 H. Liu, Y. Shen, B. Du, Z. Fang and Y. Wu, J. Mater. Chem., 2009,
1
9, 1305–1313.
3 L. Xu, Z. Guo, Y. Zhang and Z. Fang, J. Mater. Chem., 2008, 18,
083–5091.
5
14 K. Y. Mya, C. He, J. Huang, Y. Xiao, J. Dai and Y. P. Siow, J. Polym.
Sci., Part A: Polym. Chem., 2004, 42, 3490–3503.
1
5 X. Wang, Y. Hu, L. Song, W. Xing and H. Lu, J. Polym. Sci., Part B:
Polym. Phys., 2010, 48, 693–705.
(
DBP) repeating unit structure. Subsequently, the p–p stacking
3
interactions between polymeric chains drove the chains to
arrange into nanorods.
16 E. Franchini, J. Galy, J. F. G ꢁe rard, D. Tabuani and A. Medici,
Polym. Degrad. Stab., 2009, 94, 1728–1736.
1
7 T. Nazir, A. Afzal, H. M. Siddiqi, Z. Ahmad and M. Dumon, Prog.
Org. Coat., 2010, 69, 100–106.
This feature of AOPH-NR induced special properties and
performance in corresponding EP/AOPH-NR nanocomposites.
SEM and TEM images indicated the good dispersibility of AOPH-
NR in epoxy resin and enhanced mechanical properties, which was
further evidenced by DMA measurements. TG measurements
showed that the presence of AOPH-NR facilitate degradation,
which is indicative of the favorable release and migration ability of
phosphorus species for the EP/AOPH-NR. Such behavior mainly
resulted from the large surface area, good dispersion and nano-
scale of AOPH-NR. LOI measurements and cone calorimeter
analysis showed that the loading of only 4.25 wt% AOPH-NR
leads to a LOI increase ratio as high as 26.1% and a 23% reduction
in PHRR, suggesting a remarkably high flame retarding efficiency.
Relative to other flame-retardant materials, the hybrids of ATH
and phosphorus compounds offer the most impressive advantages
in tailoring releasing and migration behavior of phosphorus
species to accommodate specific requirements. This is because of
their unique one-dimensional nanorod morphology, good
dispersion, and improved phosphorus degradation temperature.
The application of this aluminum–organophosphorus hybridiza-
tion in real-world flame retardancy applications is likely.
1
8 H. Nishida, Y. Fan, T. Mori, N. Oyagi, Y. Shirai and T. Endo, Ind.
Eng. Chem. Res., 2005, 44, 1433–1437.
1
2
9 C. Jiao and X. Chen, Polym.-Plast. Technol. Eng., 2009, 48, 665–670.
0 U. Hippi, J. Mattila, M. Korhonen and J. Sepp €a l €a , Polymer, 2003, 44,
1
21 S. M. B. Nachtigall, M. Miotto, E. E. Schneider, R. S. Mauler and
193–1201.
M. M. Camargo Forte, Eur. Polym. J., 2006, 42, 990–999.
2 W. Xiao, P. He, G. Hu and B. He, J. Fire Sci., 2001, 19, 369.
3 L. Delfosse, C. Baillet, A. Brault and D. Brault, Polym. Degrad. Stab.,
2
2
1
989, 23, 337–347.
24 H. Yan, X. Zhang, L. Wei, X. Liu and B. Xu, Powder Technol., 2009,
93, 125–129.
5 T. Kanno, H. Yanase, Y. Sugata and K. Shigehara, Polym. J., 2007,
9, 347–358.
1
2
2
2
2
2
3
3
6 J. C. Liu, J. H. Jean and C. C. Li, J. Am. Ceram. Soc., 2006, 89, 882–
887.
7 C. Manzi-Nshuti, D. Wang, J. M. Hossenlopp and C. A. Wilkie, J.
Mater. Chem., 2008, 18, 3091–3102.
8 Y. N. Chan, T. Y. Juang, Y. L. Liao, S. A. Dai and J. J. Lin, Polymer,
2008, 49, 4796–4801.
9 Y. Shi, T. Kashiwagi, R. N. Walters, J. W. Gilman, R. E. Lyon and
D. Y. Sogah, Polymer, 2009, 50, 3478–3487.
0 M. Zammarano, M. Franceschi, S. Bellayer, J. W. Gilman and
S. Meriani, Polymer, 2005, 46, 9314–9328.
31 G. A. Wang, C. C. Wang and C. Y. Chen, Polym. Degrad. Stab., 2006,
1, 2443–2450.
9
3
2 D. G. Wang, F. Guo, J. F. Chen, L. Shao, H. Liu and Z. T. Zhang,
Colloids Surf., A, 2007, 293, 201–209.
Acknowledgements
3
3
3 M. Hussain and G. Simon, J. Mater. Sci. Lett., 2003, 22, 1471–1475.
4 M. Hussain, R. Varley, Z. Mathys, Y. Cheng and G. Simon, J. Appl.
Polym. Sci., 2004, 91, 1233–1253.
This work was supported through grant from the National
Natural Science Foundation of China-NSAF (Grant No
35 L. Torre, G. Lelli and J. Kenny, J. Appl. Polym. Sci., 2004, 94, 1676–
689.
3
1
1
0976024), China-NSFC (Grant No 51107106) and Open Project
6 W. Liu, R. J. Varley and G. P. Simon, Polymer, 2007, 48, 2345–2354.
7 A. Toldy, N. Toth, P. Anna, G. Keglevich, K. Kiss and G. Marosi,
Polym. Adv. Technol., 2006, 17, 778–781.
of State Key Laboratory Cultivation Base for Nonmetal
Composites and Functional Materials of Southwest University
of Science and Technology (No.11zxfk26). We thank Dr Wang
Wenhai (Yantai Institute of Coastal Zone Research, Chinese
Academy of Sciences) for SEM measurements.
3
38 B. Schartel, U. Knoll, A. Hartwig and D. P u€ tz, Polym. Adv. Technol.,
006, 17, 281–293.
2
9 A. Hartwig, D. P u€ tz, B. Schartel, M. Bartholmai and M. Wendschuh-
Josties, Macromol. Chem. Phys., 2003, 204, 2247–2257.
0 Z. Florjanczyk, A. Lasota, A. Wolak and J. Zachara, Chem. Mater.,
2006, 18, 1995–2003.
1 G. B. Hix, V. J. Carter, D. S. Wragg, R. E. Morris and P. A. Wright,
3
4
4
Notes and references
J. Mater. Chem., 1999, 9, 179–185.
42 N. Ivanova, A. Reutskaya, S. Arbuzova, L. Baikalova, A. Afonin,
1
A. B. Morgan, J. M. Cogen, R. S. Opperman and J. D. Harris, Fire
Mater., 2007, 31, 387–410.
N. Gusarova and B. Trofimov, Chem. Heterocycl. Compd., 2000,
36, 221–222.
2
3
L. Du and B. Qu, J. Mater. Chem., 2006, 16, 1549–1554.
C. Nyambo, P. Songtipya, E. Manias, M. M. Jimenez-Gasco and
C. A. Wilkie, J. Mater. Chem., 2008, 18, 4827–4838.
H. Liu, R. Wang and X. Xu, Polym. Degrad. Stab., 2010, 95, 1466–
43 M. Bujard, V. Gouverneur and C. Mioskowski, J. Org. Chem., 1999,
64, 2119–2123.
44 G. Guerrero, P. H. Mutin and A. Vioux, J. Mater. Chem., 2001, 11,
3161–3165.
45 E. A. Boyd, A. C. Regan and K. James, Tetrahedron Lett., 1992, 33,
813–816.
4
5
1
470.
H. Yukutake, M. Kobayashi, H. Otsuka and A. Takahara, Polym. J.,
009, 41, 555–561.
2
2
016 | J. Mater. Chem., 2012, 22, 2007–2017
This journal is ª The Royal Society of Chemistry 2012

View Article Online
4
6 J. M. Stinson and W. E. Horn, Journal of Vinyl Technology, 1995, 1,
64 S. Lim, K. Zeng and C. He, Mater. Sci. Eng., A, 2010, 527, 5670–5676.
ꢂ
65 X. Fern ꢁa ndez-Francos, D. Foix, A. Serra, J. M. Salla and X. Ramis,
94–97.
4
4
7 C. Misra and A. J. Perrotta, US Patent 5, 075, 472, 1991.
8 L. Le-ping, C. Xue-min, Q. Shu-heng, Y. Jun-li and Z. Lin, Appl. Clay
Sci., 2010, 50, 600–603.
React. Funct. Polym., 2010, 70, 798–806.
66 A. Omrani and A. A. Rostami, Mater. Sci. Eng., A, 2009, 517, 185–
190.
4
9 S. T. Wilson, B. M. Lok, C. A. Messina, T. R. Cannan and
E. M. Flanigen, J. Am. Chem. Soc., 1983, 218, 79–106.
0 J. Randon, P. Blanc and R. Paterson, J. Membr. Sci., 1995, 98, 119–
67 I. M. Joni, T. Nishiwaki, K. Okuyama, S. Isoi and R. Kuribayashi,
Powder Technol., 2010, 204, 145–153.
68 A. Bansal, H. Yang, C. Li, K. Cho, B. C. Benicewicz, S. K. Kumar
and L. S. Schadler, Nat. Mater., 2005, 4, 693–698.
69 P. Rittigstein, R. D. Priestley, L. J. Broadbelt and J. M. Torkelson,
Nat. Mater., 2007, 6, 278–282.
5
129.
5
5
1 P. Persson and E. Laiti, J. Colloid Interface Sci., 1997, 190, 341–349.
2 X. H. Guan, Q. Liu, G. H. Chen and C. Shang, J. Colloid Interface
Sci., 2005, 289, 319–327.
70 L. An, Y. Pan, X. Shen, H. Lu and Y. Yang, J. Mater. Chem., 2008,
18, 4928–4941.
5
5
5
5
5
5
5
6
6
6
6
3 E. Podstawka, A. Kudelski, P. Kafarski and L. Proniewicz, Surf. Sci.,
2
007, 601, 4586–4597.
71 B. Qi, Q. Zhang, M. Bannister and Y. W. Mai, Compos. Struct., 2006,
75, 514–519.
72 B. Guo, D. Jia and C. Cai, Eur. Polym. J., 2004, 40, 1743–1748.
73 W. Xu, Z. Zhou, P. He and W. P. Pan, J. Therm. Anal. Calorim., 2004,
78, 113–124.
74 P. Hornsby, Int. Mater. Rev., 2001, 46, 199–210.
75 S. V. Levchik and E. D. Weil, Polym. Int., 2004, 53, 1901–1929.
76 C. F. Kuan, W. J. Chen, Y. L. Li, C. H. Chen, H. C. Kuan and
C. L. Chiang, J. Phys. Chem. Solids, 2010, 71, 539–543.
77 L. Liu, Y. Zhang, L. Li and Z. Wang, Polym. Adv. Technol., 2010, 22,
2403–2408.
4 M. Baraska, K. Chruszcz, B. Boduszek and L. M. Proniewicz, Vib.
Spectrosc., 2003, 31, 295–311.
5 L. S. Burrell, C. T. Johnston, D. Schulze, J. Klein, J. L. White and
S. L. Hem, Vaccine, 2000, 19, 275–281.
6 J. D. Gale, A. L. Rohl, V. Milman and M. C. Warren, J. Phys. Chem.
B, 2001, 105, 10236–10242.
7 X. Ren, W. Yu, Z. Zhang, N. Xia, G. Fu, X. Lu and W. Wang,
Colloids Surf., A, 2011, 375, 156–162.
8 J. Yang, J. F. Ma, Y. Y. Liu, J. C. Ma and S. R. Batten, Cryst. Growth
Des., 2009, 9, 1894–1911.
9 W. K. Dong, C. Y. Zhao, Y. X. Sun, X. L. Tang and X. N. He, Inorg.
Chem. Commun., 2009, 12, 234–236.
0 Q. X. Liu, L. N. Yin and J. C. Feng, J. Organomet. Chem., 2007, 692,
78 H. Qin, S. Zhang, C. Zhao, G. Hu and M. Yang, Polymer, 2005, 46,
8386–8395.
79 K. Shanmuganathan, S. Deodhar, N. A. Dembsey, Q. Fan and
P. K. Patra, Polym. Eng. Sci., 2008, 48, 662–675.
80 H. Gui, X. Zhang, Y. Liu, W. Dong, Q. Wang, J. Gao, Z. Song, J. Lai
and J. Qiao, Compos. Sci. Technol., 2007, 67, 974–980.
81 H. Dong, Z. Du, Y. Zhao and D. Zhou, Powder Technol., 2010, 198,
325–329.
3
655–3663.
1 B. H. Ye, M. L. Tong and X. M. Chen, Coord. Chem. Rev., 2005, 249,
45–565.
2 J. Yang, J. F. Ma, Y. Y. Liu, J. C. Ma and S. R. Batten, Inorg. Chem.,
007, 46, 6542–6555.
5
2
3 X. M. Lin, H. C. Fang, Z. Y. Zhou, L. Chen, J. W. Zhao, S. Z. Zhu
and Y. P. Cai, CrystEngComm, 2009, 11, 847–854.
82 B. Schartel, U. Knoll, A. Hartwig and D. P u€ tz, Polym. Adv. Technol.,
2006, 17, 281–293.
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 2007–2017 | 2017