Waterborne polyurethane conjugated with novel diol chain-extender bearing cyclic phosphoramidate lateral group: Synthesis, flammability and thermal degradation mechanism

Article abstract of DOI:10.1039/c6ra06856a

A diol bearing a cyclic phosphoramidate pendant group, namely 2-(5,5-dimethyl-2-oxo-2λ⁵-1,3,2-dioxaphosphinan-2-ylamino)-2-methyl-propane-1,3-diol (PNMPD) was synthesized and characterized by FTIR and NMR. Subsequently, PNMPD was applied as a flame-retardant chain extender to prepare phosphorus-nitrogen containing waterborne polyurethane (PNWPU), and the influence of PNMPD on the hydrolysis, thermal and flame-retardant properties of PNWPU was investigated. The fabricated PNMPD-based PNWPU shows controllable mechanical properties and maintains the good hydrolysis-resistance property of polyurethane. Thermal stability and flammability analysis demonstrate that though the covalent conjugation of PNMPD induces a slight thermal destabilization effect, it efficiently promotes char formation in PNWPU, as a result, a 27.2% limiting oxygen index (LOI) value and a UL-94 V-0 rating can be achieved with only 12 wt% PNMPD incorporated. Compared with WPU, the peak heat release rate (PHRR), total heat release (THR), peak smoke production rate (PSPR) and total smoke release (TSR) of PNWPU-12, evaluated with cone calorimetry, are decreased by 44.7%, 39.0%, 42.9% and 36.1%, respectively. In addition, the flame retarding mechanism of PNWPU was comprehensively investigated by SEM microscopy, EDX analysis, real time FTIR and TG-FTIR, and results elucidate that these notable reductions in fire hazards are probably attributed to the formation of polyphosphoric acid-rich rugged and intumescent char in the condensed phase, which behaves as a "labyrinth effect" to effectively inhibit the transmission of heat, oxygen and volatile fragments from entering into the flame zone and shield the underlying PNWPU matrix against flame.

Full text of DOI:10.1039/c6ra06856a

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Waterborne polyurethane conjugated with novel
diol chain-extender bearing cyclic
Cite this: RSC Adv., 2016, 6, 56610
phosphoramidate lateral group: synthesis,
flammability and thermal degradation mechanism†
a
a
a
Peikun Zhang, Zhenyu Zhang, Haojun Fan, Saiqi Tian,a Yi Chenb and Jun Yanb
*
A
diol bearing a
cyclic phosphoramidate pendant group, namely 2-(5,5-dimethyl-2-oxo-2l5-1,3,2-
dioxaphosphinan-2-ylamino)-2-methyl-propane-1,3-diol (PNMPD) was synthesized and characterized by
FTIR and NMR. Subsequently, PNMPD was applied as a flame-retardant chain extender to prepare
phosphorus–nitrogen containing waterborne polyurethane (PNWPU), and the influence of PNMPD on the
hydrolysis, thermal and flame-retardant properties of PNWPU was investigated. The fabricated PNMPD-
based PNWPU shows controllable mechanical properties and maintains the good hydrolysis-resistance
property of polyurethane. Thermal stability and flammability analysis demonstrate that though the covalent
conjugation of PNMPD induces a slight thermal destabilization effect, it efficiently promotes char formation
in PNWPU, as a result, a 27.2% limiting oxygen index (LOI) value and a UL-94 V-0 rating can be achieved
with only 12 wt% PNMPD incorporated. Compared with WPU, the peak heat release rate (PHRR), total heat
release (THR), peak smoke production rate (PSPR) and total smoke release (TSR) of PNWPU-12, evaluated
with cone calorimetry, are decreased by 44.7%, 39.0%, 42.9% and 36.1%, respectively. In addition, the flame
retarding mechanism of PNWPU was comprehensively investigated by SEM microscopy, EDX analysis, real
time FTIR and TG-FTIR, and results elucidate that these notable reductions in fire hazards are probably
attributed to the formation of polyphosphoric acid-rich rugged and intumescent char in the condensed
phase, which behaves as a “labyrinth effect” to effectively inhibit the transmission of heat, oxygen and
volatile fragments from entering into the flame zone and shield the underlying PNWPU matrix against flame.
Received 15th March 2016
Accepted 7th June 2016
DOI: 10.1039/c6ra06856a
www.rsc.org/advances
potentially carcinogenic dioxins combined with corrosive fumes
during combustion.5,6 Consequently, tremendous research
1. Introduction
interests have been triggered in pursuit of effective halogen-free
ame-retardants, among which the phosphorus and nitrogen
(P–N) containing compounds are considered to be a promising
alternative due to their perceived ame retardant efficiency,
nontoxicity, eco-friendly and other prominent advantages.3,5–8
Generally, the P–N type ame retardants are incorporated
into polyurethane by physically blending, which is recognized
as the most convenient way.7–10 Nevertheless, this method
displays the drawbacks of considerable loading (10–30 wt%),
being susceptible to leach, poor compatibility, and deteriora-
tion in mechanical properties.5,11,12 In this regard, incorporation
of ame retardants into polyurethane backbone or side chains
via chemical covalent linkages so as to impart homogeneous
distribution, furnish permanent ame retardancy, and main-
tain the mechanical strength of polyurethane have been exten-
sively investigated.12–15 Despite the aforementioned advantages,
these corresponding reported P–N type waterborne poly-
urethanes are mainly with phosphoester-bond conjugated in
the main chain, thus they are not commercially feasible owing
to the easily hydrolytic phosphate ester linkages.5,16 The
Waterborne polyurethane (WPU) coatings have found extensive
commercial applications in various substrates such as genuine
leather, articial leather, textile, plastics, wood and ceramics by
virtue of their environment-friendly characteristics, outstanding
performances and remarkable versatility.1,2 Unfortunately, the
intrinsically high ammability of WPU has caused plenty of re-
related events with heavy casualties and tremendous property
losses over the last few years.3,4 Hence, the development of ame
retardant WPU is urgently required.
For a long time, the halogenated ingredients have undeni-
ably seized a dominant position in endowing polyurethane with
enhanced ame retardancy. However, the major problems
encountered with these species are the generation of toxic and
aKey Laboratory of Leather Chemistry and Engineering (Sichuan University), Ministry
of Education, Chengdu 610065, P. R. China. E-mail: fanhaojun@163.com; Fax: +86-
28-85401068; Tel: +86-28-85401068
bNational Engineering Laboratory for Clean Technology of Leather Manufacture,
Sichuan University, Chengdu 610065, P. R. China
† Electronic supplementary information (ESI) available. See DOI:
10.1039/c6ra06856a
56610 | RSC Adv., 2016, 6, 56610–56622
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
RSC Advances
hydrolization of organophosphate moieties will come at the solvents were commercially available in Sinopharm Chemical
expense of the scission of polyurethane backbone, which Reagent Co., Ltd. (China) and used as received.
further results in the reduction of not only ame-retardant
2.2 Synthesis of 2-(5,5-dimethyl-2-oxo-2l5-1,3,2-
efficiency but also mechanical properties. To address this
issue, one way out is to redesign reactive diol linking with P–N-
containing moieties lateral chain. Unfortunately, up to now,
rational design of such “ideal” reactive ame retardants with
pendant phosphorus and nitrogen containing group are still
scanty in the literature. In our previous work, we synthesized
dioxaphosphinan-2-ylamino)-2-methyl-propane-1,3-diol
(PNMPD)
The intermediate, 2-chloro-5,5-dimethyl-1,3,2-dioxaphosphinane-
2-oxide (DOPC) was synthesized by the reaction of POCl3 and NPG
in the presence of chloroform, according to our previously pub-
lished procedure.17 FTIR (KBr) nmax (cmꢁ1): 2976, 2946, 2896
(–CH3/–CH2); 1304 (P]O); 1052, 1005 (P–O–C); 550 (P–Cl).
PNMPD was synthesized by nucleophilic substitution reac-
tion of AMPD with DOPC as depicted in Scheme 1. Typically,
a 250 mL four-necked round-bottom ask equipped with a ther-
mometer, dropping funnel, mechanical stirrer and condenser
connected to a tail gas absorber was charged with a mixture of
AMPD (32.6 g, 0.31 mol), acetonitrile (100 mL) and triethylamine
(30.4 g, 0.30 mol). The mixture was aggressively stirred at 50 ꢀC
until AMPD completely dispersed, aer which the DOPC (55.4 g,
0.30 mol) dissolved in acetonitrile was carefully added dropwise
to the solution within a period of 2 h. Subsequently, the reaction
mixture was heated slowly up to 82 ꢀC and kept reuxing for 6 h.
Aer the solvent was removed by vacuum distillation, the
resulting crude product was repeatedly rinsed with ethyl acetate
and chloroform. Successively, the dark amber liquid, namely
PNMPD, was dried for overnight to obtain a pure product having
a yield of 86%. FTIR (KBr) nmax (cmꢁ1): 3382 (–OH and N–H);
2965, 2886 (–CH3/–CH2); 1623 (N–H); 1396 (–C(CH3)2–); 1213
(P]O); 1062 (C–O and P–O–C); 1012, 948 (P–N). 1H NMR (DMSO-
d6, 500 MHz): d ppm ¼ 0.89–1.20 (m, 9H, –CH3); 3.04 (d, 4H,
N–C–CH2); 3.75 (d, 4H, P–O–CH2); 4.72 (s, 2H, –OH); 7.93 (s, 1H,
P–NH–). 31P NMR (DMSO-d6): d ppm ¼ ꢁ6.49.
a
pendent organophosphorus-containing chain-extender
(EPPD) and conjugated it into WPU backbone to acquire
ame-retardant WPU (PWPU). An LOI value of 26.6% was ach-
ieved when the EPPD content was 12 wt%, and the hydrolytic
stability was improved compared to that with phosphoester
bond incorporated in the main-chain structures.17 Despite that,
the ame-retardant efficiency of EPPD was not very high due to
it contains only phosphorus, and the hydrolysis resistance of
PWPU still decreased a little in comparison with neat WPU.
Hence, it is necessary to develop some more efficient reactive-
type phosphorus-containing derivatives with outstanding
hydrolysis stability to ame retard WPU.
Previous research demonstrated that P–C and P–N bonds
containing monomers and polymers have preferable resistant to
hydrolysis than P–O linkages.18,19 In addition, there is synergistic
ame retardancy effect of phosphorus and nitrogen containing
compounds on polymer matrices.13,20 With this inspiration,
herein, a novel halogen-free diol bearing phosphoramidate-
containing moieties lateral chain, named as 2-(5,5-dimethyl-2-
oxo-2l5-1,3,2-dioxaphosphinan-2-ylamino)-2-methyl-propane-
1,3-diol (PNMPD), was designed and synthesized, then it was
covalently conjugated into polyurethane in the chain-extension
process to obtain a series of P–N-containing re-resistant
waterborne polyurethanes (PNWPU). A complementary study
on the ame-retardant characteristics and aming mechanism
of this polyurethane resin were also investigated. The substitu-
tion of a P–O bond with P–N bond is expected to improve the
ame resistance of PNWPU, and simultaneously, to minimize
the hydrolysis of phosphorus-containing ame retardant.
2.3 Preparation of PNMPD-based waterborne polyurethane
(PNWPU) dispersions
The ame-retardant waterborne polyurethanes bearing phos-
phorus–nitrogen moieties lateral chain were synthesized via
prepolymer method. The sample designation and theoretical
composition are tabulated in Table 1, and the polymerization
process and synthesis route of PNWPU dispersions are provided
in Fig. S1.† Taking PNWPU-6 for example, 30 g PPG220, 22 g
H12MDI and 2.9 g DMPA were charged into a 250 mL four-
necked round-bottom ask equipped with mechanical stirrer,
thermometer, nitrogen inlet and condenser, and allowed to
react at 80 ꢀC for approximately 3 h in the presence of Sn(Oct)2
as catalyst. Aerwards, 3.65 g PNMPD and 1.5 g NPG that dis-
solved in 20 mL butanone (MEK) were added to the beaker with
moderate stirring and chain extended until the theoretical
2. Experimental
2.1 Materials
Polypropylene glycol with number-average molecular weight of
2000 (PPG220) and 2,2-bis(hydroxymethyl)propionic acid
(DMPA) of reagent grade were purchased from Kanto Chemical
Co., Inc. (Japan) and dehydrated under high vacuum (0.8
mmHg) at 120 ꢀC overnight before use. Dicyclohexylmethane
diisocyanate (H12MDI) of extra pure grade was commercially
obtained from Wako Pure Chemical Industries, Ltd. (Japan) and
˚
was dewatered with 4 A molecular sieves prior to synthesis. 2,2-
Dimethyl-1,3-propanediol (NPG, Koei Chemical Company, Ltd.,
Japan), 2-amino-2-methyl-1,3-propanediol (AMPD, Tokyo
Chemical Industry Co., Ltd., Japan) and stannous octoate
(Sn(Oct)2, Alfa Aesar) were reagent quality and used without
further purication. Phosphorus oxychloride (POCl3) was kindly
received from Alfa Aesar Chemical Co., Ltd. (China) and
distilled at reduced pressure before use. Other reagents and Scheme 1 Synthetic route of PNMPD.
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 56610–56622 | 56611

View Article Online
RSC Advances
Paper
Table 1 Sample designation, composition, particle size and viscosity of PNWPU dispersions
Theoretical composition (g)
Samplea
H12MDI
PPG220
DMPA
NPG
PNMPD
TEA
H2O
HSb (wt%)
DMPA (wt%)
Particle size (nm)
WPU
22
22
22
22
22
30
30
30
30
30
2.9
2.9
2.9
2.9
2.9
3.0
2.25
1.5
0.75
0
0
2.2
2.2
2.2
2.2
2.2
108
110
112
114
116
48.19
49.13
50.04
50.92
51.77
5.01
4.92
4.83
4.75
4.66
89.78
91.91
93.65
117.1
137.8
PNWPU-3
PNWPU-6
PNWPU-9
PNWPU-12
1.82
3.65
5.47
7.3
a
b
Samples here were designated as PNWPU-x, wherein x manifested the PNMPD concentration by weight of solid content in PNWPU. Hard
segment content ¼ [mass (H12MDI + DMPA + NPG + PNMPD)]/[mass (PPG220 + H12MDI + DMPA + NPG + PNMPD)].
–NCO value was reached, which was monitored intermittently
The ammability characteristics of PNWPU lms were eval-
with the standard dibutylamine titration method (ASTM D uated with a cone calorimeter device (Fire Testing Technology,
1638). Occasionally, MEK was added to adjust the viscosity of UK) in accordance with the procedures in ISO5660. The samples
the prepolymer. Aer cooling the system to 50 ꢀC, the carboxylic with a gauge size of 100 ꢂ 100 ꢂ 2 mm3 (length ꢂ width ꢂ
groups of DMPA were neutralized with 2.2 g TEA for 15 min, and thickness) were wrapped in an aluminum foil and exposed
then the reaction mixture was dispersed by deionized water horizontally at an external heat ux of 35 kW mꢁ2
under vigorous stirring. Finally, the solvent was removed by
vacuum distillation to obtain the transparent PNWPU emul- burned PNWPU samples were determined by a Hitachi S-4700
.
The morphologies of char residues obtained from the
sion, in which the solid content was about 35 wt%.
scanning electron microscope (SEM) (Hitachi, Japan) at an
accelerating voltage of 5 kV with an energy-dispersive X-ray
(EDX) analyzer, and the surfaces were made electrically
conductive by gold-sputtered before observation.
2.4 Characterizations and measurements
Real-time Fourier transform infrared spectra (RT-FTIR) were
acquired on a Nicolet MAGNA-IR 750 FT-IR spectrophotometer
having a ventilated oven with a heating device. About 10 mg of
samples were ground with KBr powders and heated from
Fourier transform infrared spectroscopy (FTIR) of DOPC,
PNMPD, WPU and PNWPU was captured from a Nicolet IS10
FTIR spectrometer (Nicolet Instrument Company, USA) using
KBr pellets over the wavenumber range from 4000 to 400 cmꢁ1
ꢀ
ambient temperature to 600 C, meanwhile, the dynamic FTIR
at a resolution of 4 cmꢁ1
.
was tested over the wavenumber range of 4000 to 400 cmꢁ1
.
Nuclear magnetic resonance (1H NMR and 31P NMR) spectra
were recorded on a Bruker DRX500 MHz spectrometer (BRUKER
Component, Switzerland) operating at a probe temperature of
25 ꢀC with deuterated dimethyl sulfoxide (DMSO-d6) as solvent,
tetramethylsilane (TMS) as the internal reference and H3PO4
(85%) as the external standard.
TG-FTIR analysis was carried out on a TG 209 F1 apparatus
(NETZSCH, Germany) coupled with a Nicolet IS10 FTIR spec-
trometer (Nicolet Instrument Company, USA). The samples
(about 5 mg) were put in an alumina crucible and heated at
a raiꢀsed rate of 10 ꢀC minꢁ1 in the temperature range from 40 to
650 C under nitrogen atmosphere.
To evaluate the hydrolysis resistant ability of PNWPU, the
controlled hydrolysis condition was set as follows: under an air
circulating oven with a constant-temperature at 60 ꢀC, PNWPU
dispersions were incubated for 7 days. Then the treated solu-
tions were casted on polytetrauoroethylene plates to prepare
polyurethane lms. Here hydrolytic resistance was evaluated by
the retention ratio of tensile properties aer controlled hydro-
lysis. The mechanical properties of PNPWU lms were
measured according to ASTM D 638 standard: the dumb-bell
shaped lms with dimension of 50 ꢂ 10 mm2 (length ꢂ
width) were done in a UTM 6203 universal testing machine
(Shenzhen SANS Test Machine Co. Ltd., China) with a cross-
head speed of 10 mm minꢁ1 at ambient temperature.
Particle sizes of the PNWPU dispersions were determined
using a Malvern Model Zeta-sizer ZEN3600 light scattering
ꢀ
ultrane particle analyzer at 25 C.
Thermogravimetric analysis (TGA) was performed with
a NETZSCH TA Instruments High Resolution TG 209F1 thermal
analyzer under N2 atmosphere from 50 to 650 ꢀC at a scanning
rate of 10 ꢀC minꢁ1, where the dynamic nitrogen ow was kept
constant at 20 mL minꢁ1
.
The limiting oxygen index (LOI) was evaluated with a HC-2C
oxygen index analyzer instrument (Jiangning Analysis Instru-
ment Co., China) according to ASTM D 2863-2009 standard, in
which the size of each sample was 127 ꢂ 10 ꢂ 3 mm3 (length ꢂ
width ꢂ thickness) and the average values of ve replicated
experiments were reported.
The UL-94 vertical burning level was conducted by a CZF-II
instrument (Jiangning, China) in terms of ASTM D 3801 stan-
dard with samples xed vertically ca.300 mm above a cotton
layer, of which the specimen dimension was 130 ꢂ 12.7 ꢂ 3
3. Results and discussion
3.1 Characterization of PNMPD and PNWPU
mm3 (length ꢂ width ꢂ thickness). All specimens were Fig. 1 depicts the FTIR spectra of DOPC, PNMPD, WPU and
repeated in quintuple.
PNWPU. The typical absorption peaks for DOPC include 2976
56612 | RSC Adv., 2016, 6, 56610–56622
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
RSC Advances
cmꢁ1, 2946 cmꢁ1 and 2896 cmꢁ1 (–CH3/–CH2), 1304 cmꢁ1 (P]
O), 1052 cmꢁ1 and 1005 cmꢁ1 (P–O–C), 550 cmꢁ1 (P–Cl), which
is in good correlation with the prior literature.21 Compared with
DOPC, several characteristic absorption bands for PNMPD
appear at 3382 cmꢁ1, 1623 cmꢁ1, 1213 cmꢁ1 and 1012 cmꢁ1
,
which correspond to the stretching vibration of –OH and N–H,
N–H deformation, P]O stretching vibration and P–N bond,
respectively. Notably, the synthesis of PNMPD was carried out
by substitution reaction, during which the chlorine in DOPC
was consumed by the reaction with amino group in AMPD, thus
resulting in the disappearance of P–Cl at 550 cmꢁ1, thereby
directly signifying the successful synthesis of PNMPD. The
formation of PNMPD was also corroborated by 1H NMR and 31
P
NMR spectra, as presented in Fig. 2. The broad peak centered at
4.72 ppm is assigned to the protons of –OH, and the single peak
at 7.93 ppm is ascribed to the secondary amine proton, while
the high-eld signals at 0.89–1.20 ppm are reasonably attrib-
uted to the methyl protons. More importantly, only an intensive
singlet signal for phosphorus is observed at ꢁ6.49 ppm in the
31P NMR spectrum. Taken together, these results prove that the
PNMPD is fabricated successfully.
To validate the feasible covalently conjugation between
PNMPD and NCO-terminated prepolymer, the chemical struc-
ture of WPU together with PNWPU was systematically charac-
terized by FTIR, 1H and 31P NMR spectroscopy. In the FTIR
spectrum of WPU displayed in Fig. 1, the representative
absorption peaks appearing at 3345 cmꢁ1 (urethane N–H
stretching), 2970 cmꢁ1, 2925 cmꢁ1 and 2859 cmꢁ1 (C–H
symmetric and asymmetric stretching), 1707 cmꢁ1 (C]O
functional group), 1533 cmꢁ1 (urethane N–H bend vibration),
and 1104 cmꢁ1 (C–O–C) are detected. By comparison, in addi-
tion to the aforementioned characteristic bands corresponding
to polyurethane, some new absorption peaks emerge in the
FTIR spectrum of PNWPU. In detail, the presence of N–H
bending band from phosphamide is clearly evident at
1633 cmꢁ1, whilst the bands at 1305 cmꢁ1 and 1013 cmꢁ1 are
Fig. 2 1H NMR (a) and 31P NMR (b) spectra of PNMPD.
attributable to the vibration of P]O and P–N, respectively.
Unfortunately, as a consequence of band overlapping in the
1000–1200 cmꢁ1 region, the stretching vibration modes of
P–O–C cannot be revealed distinctly. However, the absorption
intensity of N–H stretching vibration for PNWPU is visibly
increased and shows a slightly redshi to downward wave-
number centered at 3324 cmꢁ1, presumedly owing to the extra
hydrogen bonding participated by urethane carbonyl groups
and N–H from phosphamide.22 This explication is supported by
the triing downeld of secondary amine proton from 6.98 ppm
to around 7.01 ppm with multiplet resonance signals in
consequence of the enhanced deshielding effects,23 as revealed
1
by H NMR spectrum in Fig. 3. Besides that, a new chemical
shi at 3.01 ppm as the assignment for methylene protons
adjacent to urethane linkage is determined, and the peak at 1.17
ppm is associated with the methyl protons in PNMPD. Of crit-
ical importance, the 31P NMR spectrum exhibits only a single
peak at ꢁ5.14 ppm, corresponding to the phosphorus proton
originating from the pendent group of PNMPD segment in
polyurethane. Associated with the FTIR results previously
described, it can be concluded that the PNMPD diols are inti-
mately and permanently tethered on PNWPU with phosphorus–
nitrogen moieties in lateral chain.
3.2 Hydrolytic resistance of PNWPU
The main concern for organophosphorus-containing water-
borne polyurethane is that phosphoester groups tend to
hydrolyze in emulsion during transport or storage, which leads
to the decrease of molecular weight and mechanical properties,
thus limits their use in many practical applications.16,24 Here the
Fig. 1 FTIR spectra of DOPC (a), PNMPD (b), WPU (c) and PNWPU (d). tensile strength and elongation at break of PNWPU lms along
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 56610–56622 | 56613

View Article Online
RSC Advances
Paper
attributable to the concentrated structural defects with
increasing PNMPD content.
As listed in Table 2, WPU displays a favourable hydrolysis
performance, which is reected by the high retention ratio of
more than 93% in mechanical properties. Further, it is inter-
esting to note that the retention rate for all PNWPU lms is
basically equivalent to that of WPU, indicating that the anchoring
of phosphamide in the dangling chains of polyurethane does not
deteriorate the hydrolysis-resistance property. Another evidence
for the excellent hydrolysis stability is illustrated by the triing
changes of ame retardancy aer controlled hydrolysis shown in
Table 4. Actually, during the hydrolyzation process of
phosphorus-containing waterborne polyurethane dispersions,
the predominant reaction is the cleavage of phosphate ester
linkages, followed by urethane bonds.27 Hence, without exag-
geration, the hydrolytic resistance of the PNWPU is essentially
determined by the phosphamide groups. For our designed
PNMPD monomer, one of the easily hydrolyzable P–O linkage is
well-substituted by P–N bond, which shows preferable resistant
to hydrolysis.19 On the other hand, in the PNWPU dispersions,
the phosphoramidate is just as a “core” that surrounded by the
“shells” of sixth position's hydrophobic methyl groups in NPG
and dioxaphosphorinane of PNMPD, conforming to the New-
man's “rule of six”.28 Accordingly, these moisture-sensitive
moieties are resistant to hydrolyse, which endows the pendant
phosphamide groups bearing PNWPU far better hydrolysis
stability than previously reported phosphate-based polymers.17,19
Fig. 3 1H NMR spectra of WPU (a) and PNWPU (b), and 31P NMR
spectra of PNWPU (c).
with the retention raꢀtio of mechanical properties aer hydro-
lytic treatment at 60 C for a week are given in Table 2. Quali-
tatively, with the conjugation of PNMPD ame retardant, the
ultimate tensile strength drops systematically from 15.0 MPa for
WPU to 8.2 MPa for PNWPU-12. However, the broken elonga-
tion increases rstly and then decreases. Generally, the inter-
connectivity and phase separation of the hard segment are two
main inuential factors that determine the mechanical strength
of the two-phase polyurethane.25 Structurally speaking, the
methyl and the bulky cyclothiophosphoramidate are anchoring
in the side-chain of PNMPD by means of special molecular
design. When conjugated into PNWPU, the stereohindrance
effects induced by these lateral groups in PNMPD are capable of
disrupting the crystallization of hard segments and encum-
bering the intermolecular reaction of polyurethane chains, thus
facilitates the slipping of macromolecular chains.26 Meanwhile,
the organophosphate ame retardant has an internal plastici-
zation effect on polyurethane matrix, which leads to so resins
and the decrease of tensile strength. On the other hand, the
appreciable reduce in ductility for PNWPU-12 may be
3.3 Thermal stability of PNMPD and PNWPU
The thermal stability and thermal decomposition behavior of
PNMPD and PNWPU were investigated by TGA in nitrogen
atmosphere. Typical TGA and DTG thermograms for PNMPD
and PNWPU are plotted in Fig. 4, and the detailed thermal
decomposition data are enumerated in Table 3.
Clearly, PNMPD decomposes in three main stages as shown
in Fig. 4, of which the maximum degradation temperature (Tmax
)
are around 253.1 ꢀC, 309.4 ꢀC and 487.2 ꢀC, respectively.
Furthermore, the 5 wt% weight loss temperature (Td5%) of
PNMPD is 211.7 ꢀC, while the char residues at 650 ꢀC is 16.7 wt%,
demonstrating that PNMPD is an efficient char-forming agent.
Fig. 4 shows that both WPU and PNWPU samples display
a three main stage degradation processes. According to
previous studies, the rst and second decomposition occurred
ꢀ
around 200–350 C is related to the cleavages of the urethane
Table 2 Mechanical properties of WPU and PNWPU films before and after hydrolysis
Tensile strength (MPa)
Before hydrolysis
Elongation at break (%)
Sample
Aer hydrolysis
Retention ratio (%)
Before hydrolysis
Aer hydrolysis
Retention ratio (%)
WPU
15.0 ꢃ 0.3
12.1 ꢃ 0.4
11.1 ꢃ 0.2
10.2 ꢃ 0.3
8.2 ꢃ 0.2
14.1 ꢃ 0.3
11.4 ꢃ 0.3
10.4 ꢃ 0.4
9.5 ꢃ 0.2
7.7 ꢃ 0.3
94.0
94.2
93.7
93.1
93.9
385.6 ꢃ 13.2
432.3 ꢃ 15.8
488.9 ꢃ 9.4
499.4 ꢃ 20.3
443.4 ꢃ 8.2
359.0 ꢃ 10.6
399.1 ꢃ 9.7
448.9 ꢃ 14.5
463.5 ꢃ 12.1
409.3 ꢃ 7.4
93.1
92.3
91.8
92.6
92.3
PNWPU-3
PNWPU-6
PNWPU-9
PNWPU-12
56614 | RSC Adv., 2016, 6, 56610–56622
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
RSC Advances
Fig. 4 TGA (a) and DTG (b) curves of PNMPD, WPU and PNWPU under nitrogen.
Table 3 TGA data for PNMPD, WPU and PNWPU
PNWPU, and the data are summarized in Table 4. Briey, the
LOI value increases monotonically from 18.5% for WPU to
27.2% for PNWPU-12, proving the enhanced ame retardancy
effect of PNMPD in polyurethane. Direct visual observation
reveals that when subjected to the butane ame at an oxygen
concentration of 21.0 vol%, the WPU ignites instantly with
vigorous ame propagation, and molten droplet with re is
observable. This melt dripping phenomenon is undesirable due
to its propensity to spread re to neighboring ignitable mate-
rials, thereby increasing the potential re hazard. Furthermore,
as a consequence of highly ammable characteristic, the WPU
specimen burns readily accompanied with serious bending
deformation till extinguishes by violent melt-dripping behav-
iors. The burning process changes signicantly with the
anchoring of PNMPD. At an oxygen concentration of 22.4 vol%,
when PNWPU-6 sample is ignited, the ame gradually becomes
small and then extinguishes with apparent melt-dripping.
While for PNWPU-12, following ignition, the ame propagates
slowly with only dim and lethargic blaze burns the testing bar at
the oxygen concentration of 27.2 vol%, and nally self-
extinguishes without aggressive molten dripping. Moreover,
PNMPD imparts waterborne polyurethane with better re
resistance than our previously synthesized pendant
phosphorus-containing chain extender at the same loading of
ame retardant. For example, the LOI value increases from
26.6% to 27.2% when conjugated with 12 wt% ame retar-
dant.17 This declares that the pendant phosphoramidate-
containing PNMPD is capable of suppressing the molten
drops to induce carbonization efficiently and there is a syner-
gistic effect in P and N element, which would allot PNWPU with
improved ame retardancy. The virtually nonammable feature
of PNWPU is also affirmed by the vertical burning test. That is,
with 12 wt% PNMPD conjugated, the corresponding poly-
urethane can easily pass UL-94 V-0 rating whilst WPU shows no
classication. In more detail, the WPU test specimen keeps
burning autogenously to the holding clamp and burns out
completely with serious melt dripping aer rst burner ame
application. Consequently, no UL-94 rating is obtained. For
a
b
b
b
c
Sample
Td5% (ꢀC) Tmax1 (ꢀC) Tmax2 (ꢀC) Tmax3 (ꢀC) Yc (%)
PNMPD
WPU
211.7
276.9
PNWPU-3 265.9
PNWPU-6 261.9
PNWPU-9 256.6
PNWPU-12 252.8
253.1
265.5
234.7
232.3
237.6
233.9
309.4
349.7
348.2
353.3
352.4
351.9
487.2
411.8
390.6
386.5
382.9
382.4
16.7
0.71
1.63
1.86
3.46
5.44
a
Td5% is dened as the temperature at which the samples undergo 5
b
wt% weight loss. Tmax is dened as the temperature at which the
c
decomposition rate of the samples reach a maximum. Char yield at
650 ꢀC.
bonds in hard segments, while the third step at 350–480 ꢀC
corresponds to the scission of so segments.29,30 For WPU lm,
the Td5% is around 276.9 ꢀC with the Tmax at 265.5 ꢀC, 339.9 ꢀC
and 402.2 ꢀC, respectively, and its residual char at 650 ꢀC is 0.71
wt%, which reects that WPU itself has a negligible char-
forming ability. Compared with WPU, the Td5% and Tmax of all
PNWPU samples show a tendency of decrease. This may be
correlated with the relatively inferior thermal stable O]P–O
and P–O–C bond in PNMPD segment.31,32 However, the degra-
dation of phosphoester bond can promote the formation of
polyphosphoric acid or its derivatives, which will catalyze the
decomposition of polyurethane to produce additional thermally
stable phosphorous-rich carbonaceous layers at elevated
temperature.9,33 Consequently, the residue percentage at 650 ꢀC
is signicantly increased from 0.71 wt% for WPU to 5.44 wt%
for PNWPU-12. The high char yield is suggested to block the
volatilization of combustible gases and inhibit the exothermi-
city of pyrolysis reaction according to the solid phase mecha-
nism of such phosphoramidate-based ame retardant, thus
bestowing polyurethane with high ame retardancy.10,11,34
3.4 Flame retardancy of PNWPU
The LOI and UL-94 vertical burning tests were used as a quan-
titative analytical method to evaluate the ammability of
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 56610–56622 | 56615

View Article Online
RSC Advances
Paper
Table 4 LOI and UL-94 values of WPU and PNWPU specimens
Untreated
UL-94
Aer controlled hydrolysis
UL-94
Sample
LOI (%)
t1a (s)
t2b (s)
Drops
Rating
LOI (%)
t1 (s)
t2 (s)
Drops
Rating
WPU
18.5 ꢃ 0.1
19.6 ꢃ 0.2
22.4 ꢃ 0.2
25.3 ꢃ 0.1
27.2 ꢃ 0.2
52 ꢃ 5
55 ꢃ 4
18 ꢃ 3
10 ꢃ 2
3 ꢃ 1
Nonec
None
8 ꢃ 2
6 ꢃ 2
2 ꢃ 1
Yes
Yes
Yes
No
Fail
Fail
V-2
V-1
V-0
18.5 ꢃ 0.1
19.6 ꢃ 0.1
22.3 ꢃ 0.2
25.1 ꢃ 0.2
27.1 ꢃ 0.1
51 ꢃ 6
53 ꢃ 4
19 ꢃ 3
10 ꢃ 3
3 ꢃ 1
None
None
8 ꢃ 3
6 ꢃ 2
2 ꢃ 1
Yes
Yes
Yes
No
Fail
Fail
V-2
V-1
V-0
PNWPU-3
PNWPU-6
PNWPU-9
PNWPU-12
No
No
a
b
c
Time to self-extinguishing aer rst burner ame application. Time to self-extinguishing aer second burner ame application. The
specimens burn up to holding clamp and burn out completely aer rst burner ame application.
PNWPU-6, the average duration of aming combustion aer the plotted in Fig. 5, and other corresponding data are recorded in
burner ame application is 18 s and 8 s, respectively. Unfortu- Table 5. It can be observed from Fig. 5 that WPU burns rapidly
nately, aming drips ignite the cotton placed below samples, aer ignition with a THR and peak HRR (PHRR) of 71.5 MJ mꢁ2
thus the UL-94 test rating is V-2. With further increased PNMPD and 1059 kW mꢁ2, respectively. For PNWPUs, as the PNMPD
concentration to 12 wt%, PNWPU-12 exhibits better re content increases, the THR and PHRR values decrease. In detail,
performance and excellent anti-dripping behaviour, and a UL- the PHRR of PNWPU-12 is 586 kW mꢁ2, accounting for 55.3% of
94 V-0 rating is achieved.
The cone calorimeter is an efficient method to simulate reduction from 71.5 MJ mꢁ2 to 43.6 MJ mꢁ2
a developing re scenario for quantifying the combustion Based on the HRR curves, the re growth rate (FIGRA) and
that of WPU. While the THR for PNWPU-12 shows a 39.0%
.
behaviors of polymers. Herein, some meaningful parameters, maximum average rate of heat emission (MAHRE) are proposed
including heat release rate (HRR), total heat release (THR), to estimate the re spread rate and burning propensity for re
smoke production rate (SPR), total smoke release (TSR) are development under full-scale conditions.35 FIGRA is calculated
Fig. 5 HRR (a), THR (b), SPR (c) and TSR (d) versus time curves of WPU and PNWPU at an applied flux of 35 kW mꢁ2
.
56616 | RSC Adv., 2016, 6, 56610–56622
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
RSC Advances
Table 5 Cone calorimeter data for WPU and PNWPU films
Sample
WPU
PNWPU-3
PNWPU-6
PNWPU-9
PNWPU-12
TTI (s)
10 ꢃ 1
19 ꢃ 2
998 ꢃ 15
23 ꢃ 2
854 ꢃ 13
28 ꢃ 3
646 ꢃ 9
39 ꢃ 3
586 ꢃ 10
PHRR (kW mꢁ2
TTPHRR (s)
)
1059 ꢃ 27
125 ꢃ 5
115 ꢃ 3
105 ꢃ 4
110 ꢃ 5
115 ꢃ 4
THR (MJ mꢁ2
)
71.5 ꢃ 2.3
296 ꢃ 11
68.7 ꢃ 1.7
332 ꢃ 13
52.8 ꢃ 0.9
278 ꢃ 8
45.9 ꢃ 1.1
235 ꢃ 9
43.6 ꢃ 0.7
198 ꢃ 6
AHRR (kW mꢁ2
FIGRA (kW mꢁ2 sꢁ1
FPI (m2 s kWꢁ1
)
)
8.5 ꢃ 0.2
8.7 ꢃ 0.3
8.1 ꢃ 0.2
5.9 ꢃ 0.1
5.1 ꢃ 0.1
)
0.009 ꢃ 0.001
423 ꢃ 15
0.019 ꢃ 0.002
414 ꢃ 11
0.027 ꢃ 0.002
349 ꢃ 10
0.043 ꢃ 0.003
289 ꢃ 8
0.067 ꢃ 0.004
255 ꢃ 5
MARHE (kW mꢁ2
)
AEHC (MJꢁk1gꢁ1
)
22.9 ꢃ 0.3
0.129 ꢃ 0.010
482 ꢃ 16
22.7 ꢃ 0.4
0.126 ꢃ 0.009
432 ꢃ 17
23.3 ꢃ 0.3
0.107 ꢃ 0.009
393 ꢃ 11
23.0 ꢃ 0.2
0.105 ꢃ 0.008
340 ꢃ 12
22.4 ꢃ 0.1
0.099 ꢃ 0.007
308 ꢃ 9
AMLR (g s
)
)
)
TSR (m2 mꢁ2
PSPR (m2 sꢁ1
Residue (%)
0.070 ꢃ 0.006
2.0 ꢃ 0.2
0.062 ꢃ 0.005
3.5 ꢃ 0.3
0.059 ꢃ 0.005
5.6 ꢃ 0.5
0.048 ꢃ 0.004
11.1 ꢃ 0.4
0.040 ꢃ 0.004
16.6 ꢃ 0.6
by the ratio of PHRR and time to PHRR (TTPHRR) whereas follow: during combustion, the phosphorus- and nitrogen-
MAHRE is derived from the cumulative hꢁe2at emission divided containing PNMPD facilitates the formation of consolidated
by time.36,37 The FIGRA value is 8.5 kW m sꢁ1 for WPU while carbonaceous layer, which acts as a barrier to inhibit the
5.1 kW mꢁ2 sꢁ1 for PNWPU-12. The MAHRE parameter for diffusion of ammable volatile fragments and prevent the
PNWPU-12 drops by 39.7% in comparison with that of WPU, underlying polyurethane from heat and oxygen.
demonstrating that the conjugation of PNMPD slows the ame
spread potential, which provides crucial time in a real world re
scenario for safe evacuation and other emergency measures to
be taken. Besides that, re performance index (FPI), dened as
the proportion of time to ignition (TTI) to PHRR, is believed to
3.5 Flame retarding mechanism
Previous studies have preliminarily validated that the residue
chars is important to the ame retardancy of polyurethane. To
further elucidate clearly the effects of PNMPD component on
be the best single indicator for overall re hazard evaluation
the char formation of PNWPU, the microstructures as well as
posed by polyurethane. Generally, a higher FPI value suggests
chemical composition of residual chars aer cone calorimeter
the longer time to ashover, and vice versa.38 From Table 5, it is
testing were assessed by SEM and EDX spectroscopy. Fig. 6
obvious that the FPI is strikingly increased with concentrated
reports some snapshots for WPU and PNWPU chars aer cone
PNMPD, further conrming an improvement of the re safety of
calorimeter tests. As can be seen from Fig. 6a that the WPU is
phosphorus- and nitrogen-containing polyurethane.
almost burnt out, leaving limited poor residue. For the PNWPU-
Highly dense smoke during the re may severely reduce the
visibility to slow down safe evacuation or even put residents to
death directly by the toxic fumes.39 Hence, smoke suppression is
of vital important. The overlay plots of the SPR and TSR are
displayed in Fig. 5c and d. The TSR value for PNWPU-12 is 308
m2 mꢁ2, much lower than 482 m2 mꢁ2 for WPU. Moreover, the
peak smoke production release (PSPR) of PNWPU-12 is merely
0.040 m2 sꢁ1, 57% of that of WPU. Consequently, both results
illustrate that the conjugation of PNMPD has a positive effect in
inhibiting smoke production.
It is noteworthy that the TSR not only reects the release of
smoke hazards, but also discloses the ame retardancy mech-
anism. A decrease in TSR evinces that the condensed-phase
ame-retardant effect plays the dominating role in
PNWPU.33,40 This statement is supported by the indiscernible
change of average effective heat combustion (AEHC) value. That
is, if a polymer exhibits ame retarding effect mainly in the
gaseous phase, the combustion of the volatilized substances
would be incomplete, thus resulting in the decrease of both
HRR and AEHC.41,42 Another evidence to further verify the above
assertion can be found by the dramatically increased char
residues from 2.0% for WPU to 16.6% for PNWPU-12 as revealed
in Table 5, which is in accord with TGA results. On the basis of
Fig. 6 Digital photographs of residues for WPU (a), PNWPU-3 (b),
the above considerations, it might be reasonable to describe as
PNWPU-9 (c), and PNWPU-12 (d) after cone calorimeter test.
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 56610–56622 | 56617

View Article Online
RSC Advances
Paper
9 sample, the char is fragile and discrete. As the PNMPD
The EDX spectrum (see Fig. S2†) demonstrates that the atom
increases to 12%, a swollen, dense and intact char layer is percentages of oxygen and phosphorus in PNWPU-12 residues
observed, which can efficiently impede both mass and heat are 23.81% and 5.83%, respectively. The phosphorus content is
transfer. Fig. 7 presents the SEM micrographs of the residue much higher than that in PNWPU-12 lm (1.47%), illustrating
surfaces of WPU and PNWPU-12. The WPU char is susceptible that the majority of phosphate derived from PNMPD remains in
to cracking during combustion and seems rather attened in the condensed phase. Additionally, the molar ratio of O to P
high magnication (Fig. 7a and b). When the combustion approaching 4 : 1 suggests that polyphosphoric acid or its
occurs, heat and oxygen are easily transferred through the derivative exist in the char.33
fractured surface. By comparison, the surface of PNWPU-12
Condensed phase and gaseous phase are two important
shows a more swollen and intumescent char structure with aspects that signicantly contribute to comprehensive under-
lots of irregular and discontinuous bubbles of about 100 mm standing of the ame retardation mechanism of polyurethane.
diameter. Specially, the microstructure of PNWPU-12 char TG-FTIR analysis was used to investigate the pyrolysis products
consists of plenty of rugged creases (shown in Fig. 7d), as if during thermal degradation process. Fig. 8 exhibits the three-
a wrinkled “carbonaceous clothes” dressed on the unburned dimensional FTIR spectrograms of gaseous volatiles evolved
polyurethane matrix. Such morphology is pivotal in preventing during the whole thermal decomposition processes of WPU
the rapid volatilization of the inside pyrolysis products from and PNWPU-12 in N2. It can be intuitively seen that the
entering into the ame zone,43 consequently attributing to the absorption peaks of gaseous products for PNWPU-12 is iden-
improvement of ame retardancy in polyurethane.
tical to that of WPU (also see Fig. S3†). However, the absor-
bance intensity of PNWPU-12 volatilized products is much
lower than WPU (see Fig. S4 and S5†), meaning less organic
volatiles release during combustion.44 The decrease of these
ammable volatiles can lessen the “fuel” to feed back to ame
and further result in the inhibition of smoke,45 which is also
veried by the reduction of THR and TSR in the cone test. More
detailed information is acquired from the FTIR spectra of the
sample at selected typical temperature given in Fig. 9. For
WPU, a slight characteristic absorbance attributing to –OH
(3600–3850 cmꢁ1, the release of water) and CO2 (2360 cmꢁ1) are
identied when the temperature rises to 200 ꢀC. With
increasing temperature, the nitrile compounꢀds (HCN gases
and –NCO groups, 2265 cmꢁ1) release at 250 C to 380 C in
ꢀ
consequence of the decomposition of urethane groups
(–NHCOOꢁ).46 Meanwhile, some new gaseous species for
hydrocarbons (2868–2970 cmꢁ1), carbonyl compounds
(1734 cmꢁ1) and ethers (C–O–C bond, 1112 cmꢁ1
)
appear
47
aer 277 ꢀC (Td5%), and these FTIR signals are substantially
Fig. 7 SEM microphotographs of the residual chars for WPU (a and b)
and PNWPU-12 (c and d) at different magnifications.
enhanced at elevated temperature, maximizing at around
Fig. 8 Three-dimensional FTIR spectra of the gaseous volatiles evolved during the combustion of WPU (a) and PNWPU-12 (b) in N2 atmosphere.
56618 | RSC Adv., 2016, 6, 56610–56622
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
RSC Advances
Fig. 9 FTIR spectra of pyrolysis gaseous products at typical temperatures during the thermal degradation for WPU (a) and PNWPU-12 (b).
ꢀ
412 ꢀC (Tmax2) and then weakening. In the meanwhile, a small nearly hardly change bꢀelow 220 C. As the pyrolysis temper-
amount of nonammable NH3 (964 cmꢁ1 and 930 cmꢁ1) is ature increases to 280 C, all these absorptions are dimmed
detected, which acts as an inert diluent in the ame zone. In slightly, indicating that the polyurethane decomposes at this
contrast to WPU, PNWPU-12 begins to release pyrolysis prod- temperature, which is also exemplied by the TGA analysis.
ucts a slightly earlier due to the degradation of PNMPD, which With further heating to 350 C, the peaks at 3379 cmꢁ1, 1717
ꢀ
coincides with the TGA results. Note that, no new absorption cmꢁ1 and 1542 cmꢁ1 vanish while other absorptions decrease
peaks emit in the FTIR spectra of volatile products in PNWPU- dramatically, thus it is unambiguously concluded that the
12, so it can be interpreted, again, that the re retardancy primary decomposition, or more exactly, the scission of hard
mechanism mainly derives from the condensed phase.
segment occurs in this corresponding stage. Meanwhile,
Real time FTIR was employed to monitor the chemical a new absorption peak at 1626 cmꢁ1 is emerged, presumably
structure changes of polyurethane in the condensed phase representing the graphite-like structure of carbon.48 Above
during thermal oxidative degradation process in N2 atmo- 450 ꢀC, almost all the characteristic peaks disappear. This
sphere. Fig. 10 depicts the RT-FTIR spectra of WPU and observation proves the complete decomposition of the
PNWPU-12 at different degradation temperatures. From the polyurethane.
spectrum of WPU, the relative intensities of the characteristic
As for PNWPU-12, there is no distinct difference in the
absorption peaks approximately at 3379 cmꢁ1 (urethane N–H absorption bands compared to WPU, apart from the peaks at
stretching), 1717 cmꢁ1 (C]O stretching vibration), 1542 cmꢁ1 1305 cmꢁ1 and 1108 cmꢁ1 that characteristic to P]O and the
(urethane N–H bend vibration), and 1107 cmꢁ1 (C–O–C) overlapping of P–O–C and C–O–C, respectively. In Fig. 10b, it is
Fig. 10 RT-FTIR spectra of WPU (a) and PNWPU-12 (b) at different pyrolysis temperatures.
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 56610–56622 | 56619

View Article Online
RSC Advances
Paper
worth noting that the intensities of most absorption peaks
Based on the foregoing analysis and in combination with the
become weaken gradually at 255 ꢀC and nally disappear at 385 previous literature,12,29,33,48,51 a plausible thermal degradation
ꢀC, which implies that the conjugation of PNMPD catalyzes the mechanism of PNWPU is postulated and depicted schematically
thermal oxidative degradation of PNWPU at lower temperature, in Scheme 2. Firstly, the urethane groups in polyurethane main-
corresponds well to the aforesaid TGA and TG-FTIR analysis. In chains and the carboxyl of DMPA degrade and release gases
fact, this is ascribe to the earlier cleavage of O]P–O and P–O–C such as CO2 and HCN at 40–253 ꢀC. Meanwhile, the
linkages as disclosed by the dramatic diminution in the peaks phosphorous-containing PNMPD dehydrates to form phos-
at 1305 cmꢁ1 and 1108 cmꢁ1 before 300 ꢀC. Compared with phoric acid. Secondly, at 253–352 ꢀC, the phosphoric acid
WPU, PNWPU-12 exhibits some new pꢀhosphorous-related catalyzes the decomposition of polyurethane into polyols,
characteristic absorption peaks above 355 C. Specically, the isocyanates, amines, HCN and CO2. Simultaneously, phos-
weak absorption appears at 1260 cmꢁ1 is assigned to the phoric acid polymerizes to form cross-linked polyphosphoric
stretching mode of P]O bond. While the absorption band at acid and its derivatives, which serve as strong Lewis acids, and
1062 cmꢁ1 (overlapping with C–O bonds) coupled with the therefore catalyze the further degradation of polyols and
shoulder peak around 930 cmꢁ1 in nger-print region that isocyanate to form dienes at 352–382 C with the evolution of
ꢀ
deriving from symmetrical and asymmetrical stretching vibra- CO2, H2O, NH3, and tetrahydrofuran. Finally, the dienes cross-
tions of P–O–P structure manifests the formation of pyrophos- link in the presence of polyphosphoric acid to form
phate in the carbon layer.49 Indeed, the appearance of P–O–P phosphorous-rich carbonaceous char with polyaromatic struc-
structure is regarded as a testimony, demonstrating that the tures at 382–600 ꢀC. And the thermal decomposition process of
PNMPD plays a positive role in the condensed phase to accel- PNWPU bearing cyclic phosphoramidate lateral chains is
erate the formation of phosphorous-containing complex- similar to previously reported ame retardant waterborne
structured char residue during combustion.50
polyurethane with phosphorus–nitrogen moieties conjugated
in the main chain structures.12,48
Scheme 2 Proposed thermal degradation mechanism of PNWPU.
56620 | RSC Adv., 2016, 6, 56610–56622
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
RSC Advances
12 L. Gu and Y. Luo, Ind. Eng. Chem. Res., 2015, 54, 2431–2438.
13 G. Wu, J. Li, C. Chai, Z. Ge, J. Lin and Y. Luo, RSC Adv., 2015,
5, 97710–97719.
14 L. Gu, Z. Ge, M. Huang and Y. Luo, J. Appl. Polym. Sci., 2015,
132(3), 41288.
15 P. Zhang, Y. He, S. Tian, H. Fan, Y. Chen and J. Yan, Polym.
Compos., 2015, DOI: 10.1002/pc.23603.
16 M. Sato, S. Endo, Y. Araki, G. Matsuoka, S. Gyobu and
H. Takeuchi, J. Appl. Polym. Sci., 2000, 78, 1134–1138.
17 P. Zhang, S. Tian, H. Fan, Y. Chen and J. Yan, Prog. Org.
Coat., 2015, 89, 170–180.
18 J. L. Hodgson and M. L. Coote, Macromolecules, 2005, 38,
8902–8910.
19 Z. Tan, C. Wu, M. Zhang, W. Lv, J. Qiu and C. Liu, RSC Adv.,
2014, 4, 41705–41713.
20 X. Zhao, F. R. Guerrero, J. Llorca and D.-Y. Wang, ACS
Sustainable Chem. Eng., 2015, 4, 202–209.
21 Q. Li, P. Jiang, Z. Su, P. Wei, G. Wang and X. Tang, J. Appl.
Polym. Sci., 2005, 96, 854–860.
4. Conclusions
In the present work, a pendant phosphoramidate group bearing
diol abbreviated as PNMPD was designed and fabricated
successfully in good yields by nucleophilic substitution reac-
tion, which was then covalently conjugated into waterborne
polyurethane in the chain-extension process to enhance the re
retardancy. The as-prepared PNWPU exhibits excellent hydro-
lysis stability in that the hydrophobic methyl pendant groups in
NPG and dioxaphosphorinane of PNMPD can well prevent water
from penetrating and attacking the phosphamide moieties.
Simultaneously, the PNWPU-12 exhibits a high LOI value of
27.2% and a decrease of PHRR, THR and SPR by 44.7%, 39.0%
and 36.1% respectively, over WPU, illustrating the signicantly
improved re resistance and sharply ameliorated smoke
suppression. Moreover, TGA, TG-FTIR, RT-FTIR, SEM combined
with EDX analysis further validate that the conjugation of
PNMPD accelerates the degradation of PNWPU at lower
temperature to promote the formation of phosphorus-rich
residual chars without altering the composition of evolved
products, which indicates the ame-retardant action of PNWPU
is condensed phase-dominant. The rugged carbonaceous layer
acts as a physical insulation barrier for underlying polyurethane
to impede heat, oxygen, and ammable gaseous products from
permeation to the ame during combustion.
22 A. M. Kaushal, A. K. Chakraborti and A. K. Bansal, Mol.
Pharmaceutics, 2008, 5, 937–945.
23 N. E. Jacobsen, NMR spectroscopy explained: simplied theory,
applications and examples for organic chemistry and structural
biology, John Wiley & Sons, 2007.
24 P. T. Knight, K. M. Lee, H. Qin and P. T. Mather,
Biomacromolecules, 2008, 9, 2458–2467.
25 D. K. Chattopadhyay, B. Sreedhar and K. V. Raju, Ind. Eng.
Chem. Res., 2005, 44, 1772–1779.
Acknowledgements
26 K. Kojio, M. Furukawa, Y. Nonaka and S. Nakamura,
Materials, 2010, 3, 5097–5110.
27 X. Liu, K. Xu, H. Liu, H. Cai, J. Su, Z. Fu, Y. Guo and M. Chen,
Prog. Org. Coat., 2011, 72, 612–620.
The authors gratefully acknowledge the nancial support from
the National Natural Science Foundation of China (Grant No.
21576172, 51273128 and 21206096).
28 M. S. Newman, J. Am. Chem. Soc., 1950, 72, 4783–4786.
29 D. Chattopadhyay and D. C. Webster, Prog. Polym. Sci., 2009,
34, 1068–1133.
Notes and references
1 Y. Lu and R. C. Larock, Biomacromolecules, 2008, 9, 3332–
3340.
´
30 Z. S. Petrovic, Z. Zavargo, J. H. Flyn and W. J. Macknight, J.
2 H. W. Engels, H. G. Pirkl, R. Albers, R. W. Albach, J. Krause,
Appl. Polym. Sci., 1994, 51, 1087–1095.
A. Hoffmann, H. Casselmann and J. Dormish, Angew. Chem., 31 J. Ding and W. Shi, Polym. Degrad. Stab., 2004, 84, 159–165.
Int. Ed., 2013, 52, 9422–9441. 32 D. Sun and Y. Yao, Polym. Degrad. Stab., 2011, 96, 1720–1724.
3 S. Bourbigot and S. Duquesne, J. Mater. Chem., 2007, 17, 33 M.-J. Chen, C.-R. Chen, Y. Tan, J.-Q. Huang, X.-L. Wang,
2283–2300.
L. Chen and Y.-Z. Wang, Ind. Eng. Chem. Res., 2014, 53,
1160–1171.
34 D. Van Krevelen, Polymer, 1975, 16, 615–620.
4 R. Davis, Y.-C. Li, M. Gervasio, J. Luu and Y. S. Kim, ACS Appl.
Mater. Interfaces, 2015, 7, 6082–6092.
5 S.-Y. Lu and I. Hamerton, Prog. Polym. Sci., 2002, 27, 1661– 35 B. Schartel and T. R. Hull, Fire Mater., 2007, 31, 327–354.
1712.
36 H. Breulet and T. Steenhuizen, Polym. Degrad. Stab., 2005,
88, 150–158.
6 C. Hamciuc, T. Vlad-Bubulac, D. Serbezeanu, I. D. Carja,
´
E. Hamciuc, G. Lisa and V. F. Perez, RSC Adv., 2016, 6.
37 Y. Tan, Z.-B. Shao, X.-F. Chen, J.-W. Long, L. Chen and
Y.-Z. Wang, ACS Appl. Mater. Interfaces, 2015, 7, 17919–
17928.
7 H. Zhu, Z. Peng, Y. Chen, G. Li, L. Wang, Y. Tang, R. Pang,
Z. U. H. Khan and P. Wan, RSC Adv., 2014, 4, 55271–55279.
8 M. Thirumal, D. Khastgir, G. Nando, Y. Naik and 38 M. El Gouri, A. El Bachiri, S. E. Hegazi, M. Rak and A. El
N. K. Singha, Polym. Degrad. Stab., 2010, 95, 1138–1145.
Har, Polym. Degrad. Stab., 2009, 94, 2101–2106.
9 M.-J. Chen, Z.-B. Shao, X.-L. Wang, L. Chen and Y.-Z. Wang, 39 X. Chen, Y. Jiang and C. Jiao, J. Hazard. Mater., 2014, 266,
Ind. Eng. Chem. Res., 2012, 51, 9769–9776.
114–121.
10 M. Neisius, S. Liang, H. Mispreuve and S. Gaan, Ind. Eng. 40 U. Braun, B. Schartel, M. A. Fichera and C. Jager, Polym.
Chem. Res., 2013, 52, 9752–9762. Degrad. Stab., 2007, 92, 1528–1545.
11 Q. Tai, Y. Hu, R. K. Yuen, L. Song and H. Lu, J. Mater. Chem., 41 J. Feng, J. Hao, J. Du and R. Yang, Polym. Degrad. Stab., 2012,
¨
2011, 21, 6621–6627.
97, 605–614.
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 56610–56622 | 56621

View Article Online
RSC Advances
Paper
42 Z. Jiang, L. Yuan, G. Liang and A. Gu, Polym. Degrad. Stab., 47 X. Wang, Y.-T. Pan, J.-T. Wan and D.-Y. Wang, RSC Adv.,
2015, 121, 30–41. 2014, 4, 46164–46169.
43 L. Song, Y. Hu, Y. Tang, R. Zhang, Z. Chen and W. Fan, 48 G. Wu, J. Li and Y. Luo, Polym. Degrad. Stab., 2016, 123, 36–
Polym. Degrad. Stab., 2005, 87, 111–116. 46.
44 B. Yu, Y. Tao, L. Liu, Y. Shi, H. Yang, G. Jie, S. Lo, Q. Tai, 49 M. Bugajny, S. Bourbigot, M. Le Bras and R. Delobel, Polym.
L. Song and Y. Hu, RSC Adv., 2015, 5, 75254–75262. Int., 1999, 48, 264–270.
45 Y. Dong, Z. Gui, Y. Hu, Y. Wu and S. Jiang, J. Hazard. Mater., 50 H. Liu, X. Wang and D. Wu, Polym. Degrad. Stab., 2014, 103,
2012, 209, 34–39. 96–112.
46 H. Singh and A. Jain, J. Appl. Polym. Sci., 2009, 111, 1115– 51 M. M. Velencoso, M. J. Ramos, R. Klein, A. De Lucas and
1143.
J. F. Rodriguez, Polym. Degrad. Stab., 2014, 101, 40–51.
56622 | RSC Adv., 2016, 6, 56610–56622
This journal is © The Royal Society of Chemistry 2016