A critical analysis of the a, b and g phases in
poly(vinylidene fluoride) using FTIR†
Xiaomei Cai,
a
Tingping Lei,
*
bc
Daoheng Sun
d
and Liwei Lin
de
Poly(vinylidene fluoride) (PVDF) has been widely utilized in scientific research and the manufacturing
industry for its unique piezoelectric properties. In the past few decades, the vibrational spectra of PVDF
polymorphic polymers via FTIR (Fourier transform infrared spectroscopy) have been extensively
investigated and documented. However, reports on the analysis of a, b and g phases often have
conflicting views based on measured data. In this work, we analyze the FTIR vibrational bands of PVDF
materials fabricated by different processes with detailed XRD (X-ray diffraction) characterization to
identify the structural a, b and g phases. By examining the results in this work and extensively reviewing
published research reports in the literature, a universal phase identification procedure using only the FTIR
results is proposed and validated. Specifically, this procedure can differentiate the three phases by
checking the bands around 763 and/or 614, 1275, and 1234 cm
1
for the a, b and g phases, respectively.
The rule for assignment of the 840* and 510* cm
1
bands is provided for the first time and an integrated
quantification methodology for individual b and g phase in mixed systems is also demonstrated.
Introduction
Since the discovery of piezoelectricity in poly(vinylidene uo-
ride) (PVDF),
1
strong interest has been focused on the pro-
cessing of the semi-crystalline polymer
2–6
and at least ve
distinct polymorphs, a (TGTG
0
, form II), b (TTTT, form I), g
(T
3
GT
3
G
0
, form III), d and 3 have been constructed depending on
the molecular chain conformation during the fabrication
processes. Processing parameters, such as mechanical, thermal,
electrical and chemical treatments all affect the nal PVDF
properties, including but not limited to electroactivity, dielec-
tric and mechanical properties, antifouling behavior, and
evoking cellular behavior.
7–12
Although FTIR has been widely
used in characterizing PVDF,
13–23
reports in the existing litera-
ture (sometimes by the same authors) have many conicting
characterization results based on FTIR results (Table 1), espe-
cially in the electroactive b and g phases. Two main reasons
have resulted in this divergence. First, many authors directly
assigned the FTIR bands at around 840 and 510 cm
1
to the
b and/or g phases without providing sufficient
evidence.
12,16–18,24–33
Second, several bands exclusive to the b and
g phases in the wavenumber range of 400–1500 cm
1
(or
smaller range) have seldom been taken into consideration
previously.
Furthermore, the relative amounts of the electroactive b and
g phases have been quantied in the works of Gregorio et al.
24
and Lopes et al.,
34
for samples containing only two-phase of
a and b;ora and g without considering the cases for three
phases (a, b, and g), or the b and g two-phase systems. Although
a recent report has proposed the quantication of b and g
phases system,
35
the procedure is rather complex. In the present
contribution, the procedure for the identication of a, b and g
phases using the FTIR vibrational spectrum is proposed and
demonstrated with an integrated quantication methodology
for individual b and g phase for PVDF materials made of various
a, b and g phase compositions. Both FTIR and XRD data have
been utilized to validate and identify the phases of various PVDF
polymeric systems, because the nonelectroactive a phase and
the electroactive b phase can be clearly identied by FTIR and
XRD, respectively.
4,5,20
Procedure for phase identification
By sorting out more than 100 prior publications, the FTIR
absorption peaks for the three main a, b and g polymorphs of
PVDF can be classied into three major categories: (1) common
peaks that appear in all three phases; (2) exclusive peaks that
only appear in one of the three phases; and (3) dual peaks that
could come from two different phases. In general, spectrum
peaks around 881, 1071, 1176 and 1401 cm
1
with high
a
School of Science, Jim ei University, Xiamen 361021, China
b
Fujian Key Laboratory of Special Energy Manufacturing, Huaqiao University, Xiamen
361021, China. E-mail: tplei@hqu.edu.cn
c
College of Mechanical Engineering and Automation, Huaqiao University, Xiamen
361021, China
d
School of Aerospace Engineering, Xiamen University, Xiamen 361005, China
e
Department of Mechanical Engineering, University of California, Berkeley, California
94720, USA
† Electronic supplementary information (ESI) available. See DOI:
10.1039/c7ra01267e
Cite this: RSC Adv.,2017,7,15382
Received 30th January 2017
Accepted 28th February 2017
DOI: 10.1039/c7ra01267e
rsc.li/rsc-advances
15382 | RSC Adv.,2017,7, 15382–15389
This journal is © The Royal Society of Chemistry 2017
RSC Advances
PAPER
Open Access Article. Published on 07 March 2017. Downloaded on 8/26/2022 12:32:43 PM.
This article is licensed under a
Creative Commons Attribution 3.0 Unported Licence.
View Article Online
View Journal
| View Issue
intensity were previously used to characterize the crystal phases
in some papers (see Table 1). However, peaks in the range of
876–885, 1067–1075, 1171–1182 and 1398–1404 cm
1
have
similar characteristics in samples of a-, b- and g-phases or other
mixed systems.
19,21,22,46,65,66,73,74,100,103
In other words, these are
common peaks that appear in all three phases. The exclusive
peaks can be used to identify the corresponding crystalline
phases, whereas the dual peaks may be the results of experi-
mental uncertainty for small wavenumber shis (typically
within 2 cm
1
), or truely different phases depending on exper-
imental conditions. Specically, the exclusive peaks for the
a phase (characteristic bands of the a phase) are around at 410,
489, 532, 614, 763, 795, 854, 975, 1149, 1209, 1383 and 1423
cm
1
; the exclusive peaks for the b phase are around at 445, 473
and 1275 cm
1
; and the exclusive peaks for the g phase are
around at 431, 482, 811 and 1234 cm
1
. In contrast, the peaks in
the range of 837–841 and 508–512 cm
1
, although may appear
in many di fferent samples, the absorbance for the b- and g-
phase is much stronger than that of the a-phase. These two
peaks (below using 840* and 510* cm
1
for simplicity) can
therefore be used to characterize the electroactive b and/or g
phases (detailed in assignment of the 840* and 510* cm
1
bands). Although the b and g phases show very close peaks in
the range of 1428–1432 cm
1
, various results support that the
peaks at 1431 and 1429 cm
1
can be used as the characteristic
bands of the b and g phases, respectively.
20,80,85,104,105
The 600
cm
1
band that was previously assigned to the b-phase by some
authors,
17,38,101,106
should not be used to characterize PVDF as
this band is commonly showed up in many samples, including
a-phase ones, due to the other intensive peak around at 613
cm
1
.
11,12,16,18,20,29,31,46,51,107,108
It is noted that the 776 and 833
cm
1
bands exclusively shown in high temperature crystalliza-
tion g-phase,
18,68,109
are rather seldom observed in other g-phase
domination samples.
14,19,40,42,55,69,70,85
Since the amounts of the three phases can be in any
percentage, establishing a universal but simplied procedure to
trace these phases is therefore of great signicance. It is found
that the bands around at 763 and/or 614, 1275, and 1234 cm
1
can be consistently used to differentiate and identify the a,
b and g phases, respectively. Therefore, the procedure for the
identication of a, b and g phases can be summarized as
illustrated in Fig. 1.
Experimental
Polymer solutions
PVDF (D692, from Shanghai Sensure Chemical Co. Ltd, with
a molecular weight of 625 000) powders were dissolved in N-
methyl-2-pyrrolidinone (NMP) and acetone to prepare the
Table 1 Divergence in the assignment of some typical bands
a
Wave number
(cm
1
)
Crystalline
phase Reference
431 b 23, 36–38
g 4, 18, 22, 24, 29, 39–41
b + g 28
482 b 42–45
g 14–15, 17, 20, 22, 23 and 46
a 47–49
510 b 12, 14, 16, 17, 19, 28, 40, 50–52
g 24, 53–56
b + g 35, 39, 41, 42, 57–60
a 47 and 48
840 b 12, 16, 17, 24, 26, 27, 51,
52, 61–67
g 15, 68–72
b + g 4, 9, 15, 18, 28, 35, 39, 57,
58, 60, 73–76
881 b 77
g 48, 51 and 72
b + g 60 and 78
a + b 62
a + b + g 73
a 79
1071 b 14, 79–84
b + g 60
a + b + g 21, 22, 46, 65, 73, 74 and 85
a 47, 48, 86–88
1176 b 34, 52, 62 and 86
g 42
b + g 14 and 60
1234 b 36, 37, 89–94
g 9, 15, 19, 21, 22, 57, 65, 73,
76, 95–99
b + g 18, 42 and 58
1275 b 14–15, 19, 22, 23, 42, 73, 76,
94, 96–98 and 100
g 71 and 72
1401 b 77, 84, 86–88
g 48
b + g 60
a 79, 101 and 102
a
Note: the assignment of some other bands such as 833 and 1431 cm
1
is also in dispute.
Fig. 1 Flow diagram for the identification of a, b and g phases; 840*
and 510* represents bands in the range of 837–841 and 508–512
cm
1
, respectively; 776
#
and 833
#
reflects possible variations based on
specific processes.
This journal is © The Royal Society of Chemistry 2017 RSC Adv.,2017,7, 15382–15389 | 15383
Paper RSC Advances
Open Access Article. Published on 07 March 2017. Downloaded on 8/26/2022 12:32:43 PM.
This article is licensed under a
Creative Commons Attribution 3.0 Unported Licence.
View Article Online
polymer solutions comprising 8 and 12 wt% PVDF with various
NMP/acetone volume ratios (V
NMP
/V
acetone
). All chemicals and
solvents were used directly without further purication. Exper-
imentally, the polymer solution was obtained by continuously
stirring the mixture at 50
C for a few hours in a Paralm-sealed
frosted glass bottle until transparent, and then degassed to
remove bubbles for the membrane preparation.
Membrane preparation
PVDF nanobrous membranes were fabricated by electro-
spinning technique reported elsewhere.
110
Non-nanobrous
membranes were prepared either by casting or spin coating
onto RCA-cleaned silicon wafers at room temperature and some
of them were further treated by a stretching process. Speci-
cally, electrospinning experiments were performed in indoor
atmosphere with xed tip-to-collector distance of 10 cm and
applied voltage of 7.5 kV, while casting and spin coating
experiments were done in a clean room. All polymer
membranes were dried at room temperature or in a tempera-
ture-controlled oven for the subsequent characterizations.
Characterization techniques
Infrared spectra of the above polymer membranes and the raw
PVDF powder were taken via a Thermo Scientic Nicolet iS50
FTIR spectrometer in the range of 400–1500 cm
1
with a reso-
lution of 2 cm
1
, where the transmission mode was adopted for
most samples except the cast membranes in the ATR (attenu-
ated total reection) mode. XRD results were obtained using
a PANalytical X'Pert PRO X-ray diffractometer with Cu-K
a
radi-
ation (wavelength 0.154 nm) operated at 40 kV and 30 mA. The
samples were scanned in the 2q range of 10
to 45
with a step
interval of 0.0167
.
Results and discussion
XRD and FTIR analyses of raw PVDF powder
It is well known that in typical PVDF samples the crystalline
content is no more than 50–60%. However, there is still no
concrete information how the amorphous part of PVDF inu-
ences the XRD and IR spectra. Therefore, the discussion in this
work concentrates on the crystalline phase, especially on the
most common a, b,andg phases. As shown in the XRD charac-
terization in Fig. 2(A), the PVDF powder is mainly made of the
a phase, as evidenced by two intensive diffraction peaks at 18.4
and 20.0
and a medium peak at 26.6
, corresponding to 020, 110
and 021 reections of the monoclinic a-phase crystal, respec-
tively.
20,111–113
It is noted from Fig. 2(A) that the powder also
presents a peak at 20.6
(inset) corresponding to 110/200 reec-
tion of the orthorhombic b-phase and four weak peaks at 33.2,
35.9, 38.8, and 41.1
corresponding to 130, 200, 002, and 111
reection of the monoclinic a-phase, respectively.
94,112,114
The
corresponding FTIR spectrum shown in Fig. 2(B) is in good
agreement with the XRD results to indicate strong a-phase crystal
based on the scheme in Fig. 1. Specically, there are two intensive
peaks at 763 and 614 cm
1
(characteristics of the a phase) with
a weak peak at 1275 cm
1
exclusive for the b phase and the
absence of the peak at 1234 cm
1
exclusive for the g phase (inset).
Hence, it can be concluded that strong a-phase with some traces
of b phase coexists in the neat PVDF powder as evidenced by the
small peaks at 1275, 841 and 510 cm
1
for the b phase.
Electrospun nanobrous membranes
As demonstrated in previous reports, PVDF nanobers
produced by the electrospinning technique are in favor of
b phase due to the large mechanical elongation and strong
electrical eld during this process.
64,101,115
However, different
phases can also be constructed by tuning electrospinning
parameters as evidenced from the XRD characterizations in
Fig. 3(A) with enlarged views of two key areas of 17–22
and 35–
42
as shown in Fig. 3(B), where three types of electrospun
membranes have been constructed: (a) 12 wt% PVDF, V
NMP
/
V
acetone
¼ 9/1, under a ow rate of 60 mLh
1
and collected at
60
C (black curves); (b) 12 wt% PVDF, V
NMP
/V
acetone
¼ 9/1,
under a ow rate of 2000 mLh
1
and collected at room
temperature (red curves); and (c) 12 wt% PVDF, V
NMP
/V
acetone
¼
5/5, under a ow rate of 60 mLh
1
and collected at room
temperature (blue curves). It is found that the type (a)
membrane is mainly in a phase as indicated by the presence of
two intensive peaks at 18.4 and 20.0
with two weak peaks at
26.6 and 35.9
, similar to that of the raw PVDF powder in
Fig. 2(A), corresponding to 020, 110, 021 and 200 reections of
the monoclinic a-phase crystal.
20,111,112,114
The crystalline phase
of the type (b) membrane is mainly in g phase as XRD exhibits
a strong peak at 20.3
and two medium peaks around 18.5 and
39.0
, which are diffraction peaks on planes (110/101), (020),
and (211) of monoclinic g-phase crystal, respectively. Several
prior reports have also described the formations of g phase by
using the electrospinning process.
19,20,61,64,74
On the other hand,
the type (c) membrane shows a very strong diffraction peak at
20.6
and a weak peak at 36.3
and can be categorized as mainly
dominated in b phase.
20,111,112,114
Fig. 3(C) shows spectra results from FTIR and Fig. 3(D) shows
the enlarged views of three areas around 510, 840, and 1430
cm
1
. Specically, if one follows the scheme in Fig. 1, the black
curve for membrane (a) shows strong a phase signal as it has
intensive peaks at 763, and 614 cm
1
(characteristics of the
a phase) without clear peaks at 1276 cm
1
(exclusive for the
b phase) and 1233 cm
1
(exclusive for the g phase). The red
curve for membrane (b) shows strong g phase signal as it has
the clear peak at 1233 cm
1
without peaks at 614/763 cm
1
(exclusive for a phase) or 1276 cm
1
(exclusive for b phase).
Furthermore, one can also identify some of the characteristic
bands of the g phase at 482, 812, and 1429 cm
1
as labeled in
Fig. 3(D). The blue curve for membrane (c) shows strong b phase
signal as it has the clear peak at 1276 cm
1
(exclusive for
b phase) without peaks at 614/763 cm
1
(exclusive for a phase)
or 1233 cm
1
(exclusive for g phase). Furthermore, one can also
identify some of the characteristic bands of b phase are at 473
and 1431 cm
1
(Fig. 3(D)). It is not surprising that the 1233
cm
1
band also appears weakly since the relaxation process (b
/ g) normally occurs in the formation of electrospun PVDF
samples.
116
It is noted that characteristic bands of a, b and g
15384 | RSC Adv.,2017,7, 15382–15389 This journal is © The Royal Society of Chemistry 2017
RSC Advances Paper
Open Access Article. Published on 07 March 2017. Downloaded on 8/26/2022 12:32:43 PM.
This article is licensed under a
Creative Commons Attribution 3.0 Unported Licence.
View Article Online
phases in the 400–460 cm
1
range have many experimental
uncertainties, thus these bands have been seldom reported in
the characterizations of electrospun PVDF previously.
19,101,115
Solution-cast membranes
Unlike electrospun membranes, the crystalline phase of
membranes prepared by the solution casting method without
experiencing any mechanical stretching is closely related to the
crystallization rate of the solution. Numerous works have
demonstrated that low-temperature solution crystallization (T <
70
C) resulted in a or g phase or their mixture, depending on
the type of solvents,
13,14,22,39,69,85,96,102,112
while some other works
reported b phase under similar conditions.
18,24,28
Here, a large
number of casting experiments have been conducted by
changing the solution concentration, solvent component, and
the crystallization temperature with detailed characterizations.
The XRD patterns and FTIR spectra of membranes cast from 8
wt% PVDF solution with V
NMP
/V
acetone
at 9/1 but dried and pro-
cessed at different temperatures are shown in Fig. 4. It is found
that the samples dried at 150 (black curves) and 40
C (red curves)
are dominated by the a and g phases, respectively, as evidenced
by the corresponding two intensive diffraction peaks at 18.4 and
19.9
and a remarkable peak at 26.6
(black curve in Fig. 4(A)),
and the corresponding two dull peaks at 18.5 and 20.2
and
a remarkable peak at 39.0
(red curve in Fig. 4(A)). On the other
hand, the sample dried at 40
C and further mechanically drawn
at 80
C with a draw ratio of 4 is dominated by the b phase, as
revealed by a very strong diffraction peak at 20.6
(inset in the
blue curve in Fig. 4(A)). The draw velocity used was 3 mm min
1
and the draw ratio was the ratio of the nal and initial sample
length. Using the scheme in Fig. 1, one can also use the FTIR
spectra in Fig. 4(B) to characterize these lms. First, the black
curve in Fig. 4(B) has strong peaks at 614 and 763 cm
1
and
should be strong in a phase; the red curve has a good peak at
1232 cm
1
and should be strong in g phase; and the blue curve
has a good peak at 1275 cm
1
and should be strong in b phase –
these correspond very well with the XRD results. One can also
check some of the characteristic bands in the three curves in
Fig. 4(B) following the scheme in Fig. 1 for a, b and g phase,
respectively, with good agreements. For example, the 447, 472
and 1431 cm
1
bands all show up in the blue curve c in Fig. 4(B)
as the strong supporting evidence for the b phase crystal. It is
noted that the 1234 cm
1
band also appears as a tiny peak in this
blue curve in Fig. 4(B), since the relaxation process (b / g)oen
occurs in stretched PVDF samples.
116
Interestingly, it is also noted
that the spin-coated PVDF lms have very similar XRD and FTIR
spectra with those of the cast lms (Part I in ESI†).
Assignment of the 840* and 510* cm
1
bands
From the literature survey, the 840* and 510* cm
1
bands have
oen yielded conicting conclusions in assigning crystal
Fig. 2 XRD pattern (A) and FTIR spectrum (B) of the raw PVDF powders.
Fig. 3 XRD patterns (A) & (B) and FTIR spectra (C) & (D) of three types
of electrospun PVDF membranes: curve ‘a’ (black color) is sample
made from 12 wt% PVDF, V
NMP
/V
acetone
¼ 9/1, under a flow rate of 60
mLh
1
and collected at 60
C; curve ‘b’ (red color) is sample made from
12 wt% PVDF, V
NMP
/V
acetone
¼ 9/1, under a flow rate of 2000 mLh
1
at
room temperature; curve ‘c’ (red color) is sample made from 12 wt%
PVDF, V
NMP
/V
acetone
¼ 5/5, under a flow rate of 60 mLh
1
at room
temperature.
This journal is © The Royal Society of Chemistry 2017 RSC Adv.,2017,7, 15382–15389 | 15385
Paper RSC Advances
Open Access Article. Published on 07 March 2017. Downloaded on 8/26/2022 12:32:43 PM.
This article is licensed under a
Creative Commons Attribution 3.0 Unported Licence.
View Article Online
phases. Experimental results from this work and the scheme in
Fig. 1 are utilized to illustrate the correct assignment proce-
dures. For example, the 839 cm
1
band is observed not only in
the g-phase electrospun membrane (red curve in Fig. 3(D)), but
also in the b-phase cast membrane (blue curve c in Fig. 4(B)). If
no exclusive bands in 400–1500 cm
1
are referred as illustrated
in Fig. 1, the arbitrary assignment of the 839 cm
1
band to one
of the electroactive phase becomes the source of confusions.
To further examine these two bands, a series of experiments
have been conducted using electrospun membranes from 12
wt% PVDF under different ow rates and ratios of NMP and
acetone (V
NMP
/V
acetone
) and examined using FTIR as shown in
Fig. 5: curve ‘a’ (black) for 60 mLh
1
and V
NMP
/V
acetone
¼ 8/2;
curve ‘b’ (red) for 2000 mLh
1
and V
NMP
/V
acetone
¼ 8/2; curve
‘c’ (orange) for 60 mLh
1
and V
NMP
/V
acetone
¼ 7/3; and curve ‘ d’
(blue) for 500 mLh
1
and V
NMP
/V
acetone
¼ 7/3. The electro-
spinning was performed at room temperature with the applied
voltage of 7.5 kV and tip-to-collector distance of 10 cm. Using
the scheme in Fig. 1 and based on the exclusive bands of the
three phases, the crystalline phase of the fabricated membrane
can be characterized as: a and b (a + b), a and g (a + g), b and g
(b + g), and a, b and g (a + b + g) phases for curves ‘a’, ‘b’, ‘c’, and
‘d’, respectively. In this example, it is observed that several
bands of b phase at 445, 473 and 1431 cm
1
and the bands of g
phase at 431, 482, 812 and 1429 cm
1
, are not observed in the
samples containing a + b phases (curve ‘a’) and b + g phases
(curve ‘c’), respectively, indicating that these bands cannot be
used exclusively for b and g phase characterizations. They are
good “supporting” evidences and there are only two exclusive
bands of 1275 cm
1
for the b phase and 1234 cm
1
for the g
phase. As such, in the presence of the 1275 cm
1
band and
absence of the 1234 cm
1
band, the 840* and 510* cm
1
bands
are considered as the b phase; on the contrary, the two bands
(840* and 510* cm
1
) are considered as the g phase. If both
1275 and 1234 cm
1
bands appear together, the two bands
(840* and 510* cm
1
) are considered as both the b and g pha-
ses; in such a case by taking a higher resolution measurement
one can split the 840* or 510* cm
1
band into another two
distinct bands for the b and g phases, respectively.
17
The above
assignment of the 840* and 510* cm
1
bands is also supported
by past prior work where the 840 cm
1
band has been assigned
to b-only, g-only or b and g phases in PVDF samples manufac-
tured by different ways.
18
Integrated quantication of individual b and g phase
Since the 840* cm
1
band can be assigned to the b, g, or both
phases based on other band information, the relative fraction of
the electroactive b and g phases (F
EA
) in terms of crystalline
components in any samples, such as a sample containing only
two phases (a + b, a + g,orb + g) or three phases (a + b + g), can
be quantied as follows:
F
EA
¼
I
EA
K
840*
K
763
I
763
þ I
EA
100% (1)
where, I
EA
and I
763
are the absorbencies at 840* and 763 cm
1
,
respectively; K
840*
and K
763
are the absorption coefficients at the
respective wave numbers, whose values are 7.7 10
4
and 6.1
10
4
cm
2
mol
1
, respectively.
24
It is easy to see from eqn (1) that, when the 840* cm
1
band
is assigned to one of the electroactive phases, the F
EA
is no
doubt equivalent to the relative fraction of the corresponding
phase. Thus, below is to show how to quantify individual b and
g phases when the 840* cm
1
band is for both phases. Fig. 6
shows the FTIR spectra of electrospun membranes using 12
wt% of PVDF mixed with different V
NMP
/V
acetone
at 3/7, 4/6, 5/5,
and 6/4 with curves labeled as ‘a’ (black), ‘b’ (red), ‘c’ (blue) and
‘d’ (orange), respectively. Insets are the magnication of
frequency regions in the 1220–1280 cm
1
(le) and 465–490
cm
1
(right) range. The electrospinning was performed at room
temperature with the applied voltage of 7.5 kV, ow rate of 500
mLh
1
and tip-to-collector distance of 10 cm. It can be seen
Fig. 4 XRD patterns (A) and FTIR spectra (B) of membranes cast from 8
wt% PVDF solution with V
NMP
/V
acetone
at 9/1: curve ‘a’ (black) for the
sample dried at 150
C; curve ‘b’ (red) for the sample dried at 40
C;
curve ‘c’ (blue) for the sample dried at 40
C and drawn at 80
C with
a draw ratio of ca. 4.
Fig. 5 FTIR spectra of electrospun membranes from 12 wt% PVDF
under different flow rates and ratios of NMP and acetone (V
NMP
/
V
acetone
): curve ‘a’ (black) for 60 mLh
1
and V
NMP
/V
acetone
¼ 8/2; curve
‘b’ (red) for 2000 mLh
1
and V
NMP
/V
acetone
¼ 8/2; curve ‘c’ (orange) for
60 mLh
1
and V
NMP
/V
acetone
¼ 7/3; curve ‘d’ (blue) for 500 mLh
1
and
V
NMP
/V
acetone
¼ 7/3.
15386 | RSC Adv.,2017,7,15382–15389
This journal is © The Royal Society of Chemistry 2017
RSC Advances Paper
Open Access Article. Published on 07 March 2017. Downloaded on 8/26/2022 12:32:43 PM.
This article is licensed under a
Creative Commons Attribution 3.0 Unported Licence.
View Article Online