You are on page 1of 12

Rapid Monitoring of Bisphenol-A Diglycidyl Ether

Formation Through Byproduct Estimation: A Study by


Fourier Transform Infra-Red Spectroscopy

Jayakumar Sreenadhan,1 Chandan Guria2


1
SGS India Private Limited, Gurgaion 122015, India
2
Department of Petroleum Engineering, Indian School of Mines, Dhanbad 826004, India

Received 21 February 2011; accepted 6 May 2011


DOI 10.1002/app.34892
Published online 1 September 2011 in Wiley Online Library (wileyonlinelibrary.com).

ABSTRACT: For identification and quantification of and a-glycol substituted resin in the polymerization mix-
desired and undesired products during epoxy resin forma- ture are analyzed and quantified using FT-IR characteristic
tion, polymerization of bisphenol-A with the stoichiomet- frequency bands at 925, 3448, 1344, 773, and 3641 cm1,
ric excess of epichlorohydrin is carried out using catalytic respectively. For optimal determination of process parame-
excess of alkali. The reaction is monitored using Fourier ters on polymerization, effect of temperature and alkali
transform infra-red (FT-IR) spectroscopy technique. Major loading on epoxy polymerization is carried to maximize
undesired products are unreacted monomers (bisphenol-A the yield of diglycidal ether of bisphenol-A with simulta-
and epichlorohydrin) and byproducts (chlorine and a-gly- neous minimizing the byproducts formation. V C 2011 Wiley

col substituted diglycidal ether of bisphenol-A), whereas Periodicals, Inc. J Appl Polym Sci 123: 28942905, 2012
the desired product is only diglycidal ether of bisphenol-
A. Molar concentrations of epichlorohydrin, bisphenol-A, Key words: polycondensation; bisphenol-A diglycidyl
diglycidal ether of bisphenol-A, chlorine substituted resin, ether; epoxy resin; reaction monitoring; FT-IR

INTRODUCTION many applications.1 The chief constituent of


DGEBPA resin is
Epoxy resin is a prepolymer of two or more reactive
terminal oxirane groups, and finds useful applica-
tion in bulk structural composite, thin specialty pro-
tective coating, electrical-electronic insulation, encap-
sulation, and stereolithographic rapid prototyping.13
With wide formulation possibilities in presence of
curing agent and processing versatilities, epoxy resin
produces crosslinked or thermoset plastics, which
where n 0 for pure monomeric DGEBPA resin, and
have excellent strength, toughness, chemical resist-
n > 40 for high molecular weight DGEBPA resin.
ance, adhesive, and electrical properties.13
A high molar ratio of EP to BPA (usually 10 : 1) is
Epoxy resins are usually prepared from the mate-
used to maximize the yield of monomeric DGEBPA,
rials having two or more reactive hydrogen atoms
whereas medium to high molecular mass of
and epichlorohydrin. More than 75% of epoxy resin
DGEBPA resin is prepared using lower molar ratios
is derived from diglycidyl ether of bisphenol-A
of EP to BPA.3 A highly exothermic base catalyzed
(DGEBPA),1 and prepared by the coupling of epi-
coupling of EP and BPA proceeds via mono- and
chlorohydrin (EP) and bisphenol-A (BPA). The mar-
dichlorohydrin intermediates, which further undergo
ket dominance of epoxy resins is mainly due to its
dehydrochlorination to glycidyl ether with stoichio-
low cost and adequate-to-superior performance in
metric amount of base.2,3 Sodium hydroxide, lithium
salts, and quaternary ammonium salts are the com-
mon base catalysts for coupling reaction.4 Two-stage
Correspondence to: C. Guria (guria.c.pe@ismdhanbad.ac. reaction between EP and BPA usually occurred in
in). the presence of alkali to produce monomeric
Contract grant sponsor: All India Council for Technical DGEBPA resin. The detailed description of mono-
Education (AICTE), Government of India, New Delhi; mers and polymerization products is summarized in
contract grant number: 8023/RID/BOR/RPS-6/2005-06.
Table I. Details of main reactions and side reactions
Journal of Applied Polymer Science, Vol. 123, 28942905 (2012) are shown in Table II. Final resin properties are
V
C 2011 Wiley Periodicals, Inc. adversely affected by the purity of monomeric
RAPID MONITORING OF BISPHENOL-A DIGLYCIDYL ETHER 2895

TABLE I
Details of Monomers and Polymerization Products
Name Abbreviation Formula

Epichlorohydrin EP

Bisphenol-A BPA

Diglycidyl ether DGEBPA


of Bisphenol-A

Bis-chlorohydrin ClH-DGEBPA
of DGEBPA

Bound chlorine bCl-DGEBPA


of DGEBPA

Alpha glycol or bis-diol of a-DGEBPA


Diglycidyl ether of
Bisphenol-A (AG)

Abnormal chlorohydrin abClH-DGEBPA


of DGEBPA

DGEBPA resin, which depends on undesirable side yield of monomeric DGEBPA resin, especially in
products, intermediates, and unreacted monomers large scale production. Details of byproducts forma-
content in monomeric DGEBPA resin. Common tion are given in Table I. Dehydrochlorination of
undesirable side products are a-glycol substituted chlorohydrin intermediates produces water [Table II,
DGEBPA resin (a-DGEBPA) and chlorine substituted eqs. (2) and (3)], which results unavoidable hydroly-
DGEBPA resin (Cl-DGEBPA).3,5 Undesired side sis of epoxy ring producing hydrolyzed resin [i.e., a-
products in resin not only affect the processing char- DGEBPA: Table II, eq. (4)]. The a-DGEBPA content
acteristics and the resin purity but also reduce the in resin increases the rate of curing with diamines,

TABLE II
Main and Side Reactions for DGEBPA Formation3,5
Main reactions:
(1) BPA 2EP 2NaOH!ClH-DGEBPA 2NaOH Base catalyzed coupling
(2) ClH-DGEBPA!DGEBPA 2HCl Dehydrochlorination
(3) 2HCl 2NaOH!2NaCl 2H2O Water formation
Side reactions:
(4) DGEBPA 2H2O!a-DGEBPA a-Glycol formation
(5) ClH-DGEBPA 2EP 2NaOH!bCl-DGEBPA 2NaCl 2H2O Bound chlorine formation
(6) BPA 2EP!abClH-DGEBPA Abnormal addition to EP

Journal of Applied Polymer Science DOI 10.1002/app


2896 SREENADHAN AND GURIA

and directly affects the mechanical properties of the tio of EP to BPA), and (ii) the estimation of the side
cured resin.1 Incomplete dehydrochlorination of products during polymerization. Till date, detection
mono- and dichlorohydrin intermediates results the and quantitative estimation of undesired products
formation of bischlorohydrins of DGEBPA resin [i.e., during polymerization of EP and BPA with varying
ClH-DGEBPA: Table II, eq. (2)]. Excess EP also process conditions is not available in the open litera-
reacts with ClH-DGEBPA resin giving bound chlo- ture. Due to rapid development of reliable instru-
rine in DGEBPA resin [i.e., bCl-DGEBPA: Table II, mentation technique and computerized data process-
eq. (5)], which is responsible for chain branching in ing facilities, now-a-days, Fourier transform infra-
epoxy resin.3,5 Increased branching due to the for- red spectroscopy (FT-IR) is becoming a very power-
mation of bCl-DGEBPA resin increases the viscosity ful tool for detection and quantitative estimation of
of the reaction mixture and poses difficulties in han- compounds,16,17 and this technique is applied for
dling of resin during reaction.5 Abnormal phenoxide quantitative estimation of epoxy resin curing fre-
attack on secondary carbon atom of EP results an quently.10,18,19 Due to the well-defined characteristic
end group, which is unable to undergo dehydro- FT-IR transmittance peaks of EP and BPA polymer-
chlorination and produces abnormal chlorohydrins izing mixture10,2024 [i.e., epoxy ring (EP), glycidyl
[i.e., abClH-DGEBPA: Table II, eq. (6)]. In the pres- group (mono- and diglycidyl ether), hydroxyl group
ent study, all the chlorine substituted DGEBPA res- (BPA and a-glycol), and chlorine group (Cl-
ins (i.e., ClH-DGEBPA, bCl-DGEBPA, and abClH- DGEBPA)], FT-IR technique can be used for rapid
DGEBPA) are collectively termed as Cl-DGEBPA reaction monitoring through identification and quan-
resin for simplification. These chlorine intermediates tification during epoxy resin formation.
adversely affect the electrical properties of resin.3,5 In the present study, monomeric DGEBPA resin is
For microelectronic applications, the increased level synthesized using BPA and controlled molar excess
of total-, ionic-, and/or saponifiable-chlorine also of EP with catalytic excess of sodium hydroxide. Po-
affects the dielectric properties of the resin.5 Pres- lymerization reaction is monitored through detection
ence of chlorine in epoxy resin also produces corro- and quantification of desired product (i.e., DGEBPA
sive and obscuring smoke while burning, and gives resin), undesired byproducts (i.e., a-DGEBPA resin
supertoxic halogenated dibenzodioxines and diben- and Cl-DGEBPA resin), and unreacted monomers
zofurans, which has deleterious effects on environ- (i.e., EP and BPA) using the characteristic FT-IR
mental impact and human health.6 Unreacted BPA transmittance peaks. Effect of temperature and alkali
in epoxy resin is also toxic and migrates from can- loading on polymerization is also studied for finding
coating into food on prolonged storage that affects the optimal process parameters such that the yield
adversely human estrogenic activity.7 It has also of desired product is maximized with simultaneous
been reported that a-DGEBPA resin and Cl-DGEBPA minimization of the undesired products during
resin can attack and bind covalently DNA molecules monomeric DGEBPA resin formation.
causing mutagenic events.8,9 Thus, the presence of
undesirable products in DGEBPA resin (i.e., a-
DGEBPA, Cl-DGEBPA, and unreacted BPA) not EXPERIMENTAL
only reduces the yield of resin but also affects
Materials
adversely the resin properties in the large extent. To
improve the yield and quality of DGEBPA resin, it is Bisphenol-A, epichlorohydrin, and sodium hydrox-
necessary to identify and quantify all the undesirable ide were purified for molecular weight control of
products during DGEBPA resin formation under monomeric DGEBPA resin. Bisphenol-A (CDH, New
varying process conditions. Quantitative estimation Delhi) was purified by recrystallization from tolu-
of the byproducts is usually carried out by the ene.25 Epichlorohydrin with 98% assay (CDH, New
chemical methods of analysis which requires several Delhi) was purified by distillation under vacuum.25
hours to quantify, and difficult to apply for reaction Sodium and potassium hydroxide pellets (Sd Fine
monitoring of DGEBPA resin formation. Details of Chem Ltd, Mumbai) were dried in silica gel filled
the chemical method of analysis are reported by vacuum desiccators before use. Pyridine (Aldrich)
Lapkin and Weir,10 and Stenmark and Weiss.11 was purified after total refluxing using dry sodium
Reaction monitoring of epoxy resin formation was hydroxide pellets for 6.0 h followed by distillation.25
carried out by several investigators through the Reagent grade benzene, toluene, isopropanol, chloro-
quantification of desired product only, and the form, methanol, glacial acetic acid, hydrochloric acid
quantification is mostly based on molecular weight (AR), oxalic acid, cetyltrimethyl ammonium bro-
determination using gel permeation chromatogra- mide, quaternary ammonium periodate, potassium
phy.1215 Moreover, above studies did not account (i) iodide, silver nitrate, sulfuric acid, sodium thiosul-
the variation of process conditions (e.g., polymeriza- phate, phenolphthalein, potassium chromate were
tion temperature, alkali concentration, and molar ra- used directly for analysis without purification.

Journal of Applied Polymer Science DOI 10.1002/app


RAPID MONITORING OF BISPHENOL-A DIGLYCIDYL ETHER 2897

Synthesis of DGEBPA resin and byproducts SS 316 autoclave (Amar Equipments Pvt. Ltd., Mum-
for standardization bai).28 Excess of tetrahydrofuran was removed
Diglycidyl ether of bisphenol-A, DGEBPA resin though the chilled water circulated overhead con-
denser under vacuum and residual tetrahydrofuran
Epoxy resin (average molecular weight: 370 i.e., was removed after drying under vacuum at 313 K
monomeric DGEBPA) was synthesized using 10 : for 24.0 h. Hydrolysable chlorine in the resin (i.e., an
1M ratio of EP and BPA with catalytic excess of so- incomplete dehydrochlorination intermediate readily
dium hydroxide at 333 K in a 1.0 L semibatch SS 316 undergoes epoxy ring formation) was determined
autoclave [Table II, eqs. (1)(3)].26 The autoclave using potassium hydroxide solution and liberated
(Amar Equipments Pvt. Ltd., Mumbai) was chlorine was titrated with silver nitrate solution
equipped with automatic temperature controller using potassium chromate indicator.10 Total chlorine
(split range type with simultaneous heating and content in resin i.e., sum of hydrolysable chlorine
cooling arrangement) and overhead chilled water [Table II, eq. (2)], bound chlorine [Table II, eq. (5)],
circulated reflux condenser. Autoclave was fitted and abnormal chlorohydrin [Table II, eq. (6)] of
with a six-blade turbine and connected through a DGEBPA resin was determined by Parr bomb proce-
magnetic coupling for zero leakage. Chilled water at dure.10 In the present study, the estimated hydrolys-
283 K was circulated through cooling coil inside the able chlorine and the total chlorine were found to be
reactor to remove high exothermic heat of polymer- almost identical. Therefore, it is assumed that total
ization. After the completion of reaction, excess EP chlorine is mostly due to the hydrolysable chlorine
in the reaction mixture was removed through the only, which was found to be 12.5% (wt/wt).
overhead condenser under vacuum. Residual EP in
resin was removed under vacuum drying at 313 K
for 24.0 h. Epoxy equivalent of DGEBPA resin was FT-IR characterization of monomers, DGEBPA
determined using pyridinium chloride-pyridine resin, and byproducts
method.10 The weight per epoxy equivalent (WPE) In the present study, coupling between EP and BPA
was found to be 184.8 g of epoxy resin per gram in presence of catalytic excess of alkali was moni-
equivalent of epoxy group, which confirms the pu- tored using FT-IR technique. FT-IR spectrums were
rity of monomeric DGEBPA resin (i.e., average mo- obtained using IR Prestige 21, Shimadzu Corp., Ja-
lecular weight of resin 370). pan. Scanning conditions during spectral analysis
were maintained at (i) wave number: 4000 to 400
cm1, (ii) resolution: 4 cm1, (iii) number of scans:
a-Glycol of DGEBPA, a-DGEBPA resin
64, (iv) scan speed: 0.63, and (v) detector: Ge coated
a-DGEBPA was synthesized after hydrolysis of KBr pellet. The sample chamber was purged with
monomeric DGEBPA resin [Table II, eq. (4)].27 Hy- dry nitrogen at 120 cm3/s flow rates before analysis.
drolysis of DGEBPA was also carried out in 1.0 L Benzene was found to be suitable solvent for stand-
automatic temperature control semibatch SS 316 ardization of EP, BPA, DGEBPA, Cl-DGEBPA, and
autoclave (Amar Equipments Pvt. Ltd., Mumbai). a-DGEBPA in infra-red region.10 Following charac-
The resin was mixed with molar excess of distilled teristic infrared transmittance frequency bands of
water and refluxed at 373 K in the autoclave. Hydro- EP, BPA, DGEBPA, Cl-DGEBPA, and a-DGEBPA
lysis was continued till the exothermic reaction sub- were considered for the calibration10,2024:
sides. Time required to complete the polymerization
was almost 4.0 h. Excess of water in the reaction EP: 925 cm1 (in phase stretching frequency
mixture was removed through overhead condenser due to ring breathing vibration in EP of CAC,
under vacuum. Residual moisture was removed by CAO, and CAO bonds),
drying under vacuum at 313 K for 24.0 h. Dried mix- BPA: 3448 cm1 (asymmetric stretching vibra-
ture was analyzed for a-glycol content in the resin tion of OH group),
by Stenmark and Weiss method,11 which was found DGEBPA: 1344 cm1 (in phase stretching fre-
to be 0.825 mol/100 g of resin. quency due to the ring breathing vibration of
CAC, CAO, and CAO bonds in DGEBPA),
Cl-DGEBPA: 773 cm1 (symmetric stretching
Bis-cholorohydrin of DGEBPA, of CACl bond), and
ClH-DGEBPA resin
a-DGEBPA: 3641 cm1 (OH stretching defor-
To synthesize ClH-DGEBPA resin, monomeric mation in the nonbonded primary alcohol).
DGEBPA resin was dissolved in tetrahydrofuran
and allowed to react with concentrated hydrochloric Above characteristic infrared absorption frequency
acid (35% wt/wt) maintaining at 253 K using cryo- bands are used for detection and quantitative esti-
stat (refrigerant: glycol and water) for 6.0 h in 1.0 L mation of the above components present in the

Journal of Applied Polymer Science DOI 10.1002/app


2898 SREENADHAN AND GURIA

Figure 1 FT-IR spectra (percent absorbance versus wave number) for (a) epichlorohydrin, (b) bisphenol-A, (c) diglycidyl
ether of bisphenol-A, (d) chlorine substituted diglycidyl ether of bisphenol-A (chlorohydrins), and (e) diol substituted
diglycidyl ether of bisphenol-A (alpha glycol, AG) using standard solutions.

polymerizing mixture. It is also noted that the sym- at 773 cm1.23 Though, these frequencies are close to
metric stretching of CACl bond in EP appears at 735 each other, but the transmittance peak at 773 cm1 is
cm1 due to its stable conformer,24 whereas symmet- sufficient for identification of CACl bond in Cl-
ric stretching of CACl bond in Cl-DGEBPA appears DGEBPA resin.

Journal of Applied Polymer Science DOI 10.1002/app


RAPID MONITORING OF BISPHENOL-A DIGLYCIDYL ETHER 2899

Figure 2 Calibration plots (percent absorbance vs. molar concentration) for (a) epichlorohydrin, (b) bisphenol-A, (c)
diglycidyl ether of bisphenol-A, (d) chlorine substituted diglycidyl ether of bisphenol-A (chlorohydrins), and (e) diol sub-
stituted diglycidyl ether of bisphenol-A (alpha glycol, AG).

Calibration of monomers, DGEBPA pared using cold benzene solvent. Here, molar con-
resin, and byproducts
centrations were reported in weight basis instead of
Standard liquid samples with varying molar concen- volume basis to ensure the accuracy of the calibra-
tration (i.e., mmol/g of benzene) of EP, BPA, tion plots. First, standard FT-IR plots (i.e., percent
DGEBPA, Cl-DGEBPA, and a-DGEBPA were pre- transmittance vs. wave number) for all the standard

Journal of Applied Polymer Science DOI 10.1002/app


2900 SREENADHAN AND GURIA

TABLE III
Details of the Polymerization Experiments
Experiment no Temperature (K) EP, g (mol) BPA, g (mol) NaOH, g (mol)

1 333 231.25 (2.5) 57 (0.25) 20.5 (0.5125)


2 353 231.25 (2.5) 57 (0.25) 20.5 (0.5125)
3 363 231.25 (2.5) 57 (0.25) 20.5 (0.5125)
4 373 231.25 (2.5) 57 (0.25) 20.5 (0.5125)
5 383 231.25 (2.5) 57 (0.25) 20.5 (0.5125)
6 363 231.25 (2.5) 57 (0.25) 21.0 (0.5250)
7 363 231.25 (2.5) 57 (0.25) 22.0 (0.5500)
8 363 231.25 (2.5) 57 (0.25) 25.0 (0.6250)

samples of EP, BPA, DGEBPA, Cl-DGEBPA, and a- inside the reactor to remove high exothermic heat of
DGEBPA were generated using IR Prestige 21, Shi- reaction during epoxy resin formation. For uniform
madzu Corp., Japan. A multipoint base line correc- mixing, stirrer rotation was kept at 1000 rpm, which
tions were made for all the desired transmittance also helps to maintain uniform polymerization tem-
peaks and shifted to 100% transmittance line to perature. A typical reactor charge consisting of
avoid instrumental errors.18 Now, the absorbance 231.25 g (2.50 mol) EP, 57.00 g (0.25 mol) BPA, and
peaks for each corrected transmittance peaks were 20.50 g (0.5125 mol) NaOH was taken into the auto-
calculated and percent absorbance versus wave clave and allowed for the complete dissolution.
number plots were generated for all the standard Then, reaction temperature was raised to the desired
samples of EP, BPA, DGEBPA, Cl-DGEBPA, and a- temperature. Molar ratio of EP and BPA for resin
DGEBPA. The calibration results for EP, BPA, formation was fixed at 10 : 1 to produce monomeric
DGEBPA, Cl-DGEBPA, and a-DGEBPA are shown DGEBPA. Details of the polymerization temperature,
in Figure 1(ae), respectively. Corresponding to the EP, BPA, and alkali loadings in the present study
Figure 1(ae), calibration plots (i.e., percent absorb- are reported in Table III. Initial sample was collected
ance vs. molar concentration in mmol/g) for EP, after attaining the desired polymerization tempera-
BPA, DGEBPA, Cl-DGEBPA, and a-DGEBPA are ture and thereafter samples were collected at the
shown in Figure 2(ae), respectively. Linear relation- regular interval of time. Collected samples were
ship between percent absorbance versus molar con- weighted and quenched with known amount of cold
centration (mmol/g) for all the calibration plots are benzene at 283 K to precipitate sodium chloride [Ta-
obtained with high correlation coefficients and ble II and eq. (3)]. Sodium chloride was removed af-
shown in Figure 2(ae). Corresponding to each best ter centrifugation and filtrate was analyzed for FT-IR
fit calibration plots, best fit equations are also shown analysis. Spectral analysis of the polymerizing sam-
in Figure 2(ae) for all the standard samples. Now, ples containing BPA, DGEBPA, Cl-DGEBPA, and a-
these best fit equations are used for the determina- DGEBPA were made and the characteristic transmit-
tion of unknown molar concentrations of EP, BPA, tance peaks were found to be identical with the ref-
DGEBPA, Cl-DGEBPA, and a-DGEBPA in the poly- erence transmittance peaks. In the case of EP, char-
merizing mixture. acteristic transmittance peak is shifted from 925
cm1 to 912 cm1, and the shifting of this peak is
mainly due to hydrogen bonding between oxirane
Procedures for temperature and ring oxygen with hydroxyl compounds in BPA and/
catalyst loading study
or a-glycol, which are present in the polymerizing
Polymerization of EP and BPA in presence of cata- mixture. Shifting of this transmittance peak was also
lytic excess of NaOH was carried out with varying confirmed by analyzing FT-IR spectrum for the mix-
reaction temperature and catalyst loading for given ture of EP and isopropyl alcohol in benzene. There-
molar ratio of EP and BPA. For reaction monitoring, fore, the molar concentrations of all the components
polymerization was carried in a smaller capacity in the polymerizing mixture are directly obtained
semibatch 400 mL SS 316 autoclave (Amar Equip- from the calibration plots at 925 cm1, 3448 cm1,
ments Pvt. Ltd., Mumbai) with automatic tempera- 1344 cm1, 773 cm1, and 3641 cm1 FT-IR peaks for
ture controller to remove exothermic heat of poly- EP, BPA, DGEBPA, Cl-DGEBPA, and a-DGEBPA
merization efficiently. In this autoclave, a six-blade respectively. Usually, molar concentration of the
turbine was also fitted with the stirrer through a component present in the reaction mixture is
magnetic coupling for zero leakage. Chilled water at expressed in terms of mmol/L or mol/L of reaction
283 K was used to circulate through cooling coil mixture volume. In the present study, molar

Journal of Applied Polymer Science DOI 10.1002/app


RAPID MONITORING OF BISPHENOL-A DIGLYCIDYL ETHER 2901

concentration of the reaction mixture is reported exothermic, it is possible to maintain the desired
in mmol/L of reaction mixture. Following procedure temperature within 6 1.0 C using chilled water at
was followed to convert mmol/g of benzene (cali- 283 K with high recirculation rate, and keeping the
bration plot) to mmol/L of polymerizing mixture: stirrer rotation at 1000 rpm. Samples were collected
at the regular interval of time and analyzed using
i. mmol/g of benzene was calculated from the previously mentioned conditions of FT-IR. Molar
calibration plot [Fig. 2(ae)] based on the concentrations of EP, BPA, DGEBPA, Cl-DGEBPA,
actual absorbance value of the withdrawn and a-DGEBPA (mmol/L) were measured using the
sample, which was mixed with cold benzene. respective calibration plots [i.e., Fig. 2(ae)]. Varia-
ii. Based on the amount of benzene added (in tion of molar concentration in mmol/L of EP, BPA,
grams) for quenching, mmol of component in DGEBPA, Cl-DGEBPA, and a-DGEBPA with poly-
the polymerizing mixture was calculated. merization time at different temperatures are shown
iii. Volume of withdrawn sample (excluding ben- in Figure 3(ae) respectively.
zene) was calculated based on the weight of Figure 3(a) shows the variation of EP concentra-
sample collected and the average density of tion (mmol/L) with polymerization temperature. It
the polymerizing mixture. The average density is found that the molar concentration of EP gradu-
of the polymerizing mixture in the present ally decreases with polymerization time and conver-
study was measured and found to be 1.2473 sion of EP is enhanced with the increase in polymer-
g/cm3. ization temperature. Enhancement of EP conversion
iv. Finally, mmol/L of polymerizing mixture was at higher temperature is an indication of the strong
obtained from the number of mmol calculated temperature dependent polymerization reaction.
and the volume of polymerizing mixture with- However, the depletion of BPA is different from EP.
drawn from the autoclave. At lower polymerization temperature (i.e., 333 K,
353 K, and 363 K), molar concentration of BPA
The quantitative analysis by FT-IR technique in decreases rapidly to zero concentration level and
the present study was also confirmed by the chemi- overlaps with the X-axis [Fig. 3(b)]. Similarly, molar
cal method of analysis for the estimation of resin concentration of BPA remains nonzero at higher
byproducts, mainly, DGEBPA, Cl-DGEBPA, and a- temperature especially at 373 K and 383 K [Fig.
DGEBPA. For this, final polymerization sample at 3(b)]. Usually, phenol compounds are acidic in na-
240 min was analyzed for DGEBPA, Cl-DGEBPA, ture and results the corresponding phenoxide ions
and a-DGEBPA content using pyridinium chloride- in presence of strong alkali e.g., NaOH.29 Therefore,
pyridine method,10 Stenmark and Weiss method,11 the vanishing molar concentration of BPA at low
and Parr bomb procedure,10 respectively, and the temperature is mainly due to formation of the corre-
results were compared with the results obtained sponding phenoxide ion and results the disappear-
from FT-IR analysis. It was found that the analytical ance of OH group due to asymmetric stretching
results were almost identical with FT-IR results. vibration at 3448 cm1. It is also noted that the
Therefore, FT-IR technique can be used for the quan- molar concentration of BPA remains unchanged at
tification of EP, BPA, DGEBPA, Cl-DGEBPA, and a- 383 K, when polymerization is carried out for the
DGEBPA accurately during epoxy resin formation. longer period of time [Fig. 3(b)]. Usually, in aqueous
In the present study, monomeric DGEBPA (i.e., n solutions phenol exists in equilibrium with phenox-
0) is obtained and weight per epoxy equivalent is ide ion29 and reversible reaction (i.e., sodium phen-
184.8 g epoxy/g equiv epoxy group with average oxide to BPA) predominates at higher temperature,
molecular weight of resin is 370. Therefore, the effect which controls the disappearance of BPA during po-
of chain length on FT-IR results will be almost identi- lymerization [Fig. 3(b)]. Therefore, it is desirable to
cal for all the analysis. keep low polymerization temperature to maintain
minimum molar concentration of BPA in the resin. It
is also mentioned that the initial molar concentration
RESULTS AND DISCUSSION of EP and BPA (i.e., after attaining the desired poly-
merization temperature) is lower than that the start-
Effect of temperature on DGEBPA resin and ing individual molar concentration of EP and BPA.
byproducts formation
The depleted initial molar concentration of BPA and
Effect of polymerization temperature on the product EP is mainly due to the prepolymerization of EP
distribution including main product and undesired and BPA before attainment of the desired polymer-
products was studied, keeping constant molar ratio ization temperature.
of EP to BPA and alkali loading (Table III). Polymer- Formation of DGEBPA resin during polymeriza-
ization temperatures were fixed at 333 K, 353 K, 363 tion is shown in Figure 3(c) with varying polymer-
K, 373 K, and 383 K. Though the reaction is highly ization temperature. Here, molar concentration of

Journal of Applied Polymer Science DOI 10.1002/app


2902 SREENADHAN AND GURIA

Figure 3 Variation in molar concentration with time under varying polymerizing temperature for (a) epichlorohydrin,
(b) bisphenol-A, (c) diglycidyl ether of bisphenol-A, (d) chlorine substituted diglycidyl ether of bisphenol-A (chlorohy-
drins), and (e) diol substituted diglycidyl ether of bisphenol-A.

DGEBPA resin steadily rises with time and attains lymerization temperature. It is also noted that
almost a steady value after allowing sufficient poly- the initial molar concentration of DGEBPA resin for
merization time. As expected, molar concentration all temperature studied takes positive value and
of DGEBPA resin increases with the increase in po- this is mainly due to the formation of DGEBPA

Journal of Applied Polymer Science DOI 10.1002/app


RAPID MONITORING OF BISPHENOL-A DIGLYCIDYL ETHER 2903

resin takes place before attending specified reaction increases the yield of DGEBPA resin, Cl-DGEBPA
temperature. resin, and a-DGEBPA resin (except a-DGEBPA resin
The variation in molar concentration of Cl-DGEBPA formation at 333 K), whereas conversion BPA is
resin with polymerization temperature is shown in reduced with the increase in temperature for given
Figure 3(d). It is observed that molar concentration of alkali loading.
Cl-DGEBPA resin increases at the beginning of poly-
merization and falls to a minimum value at the end
of polymerization after attaining the maximum peak. Effect of alkali concentration on DGEBPA resin
and byproducts formation
The typical concentration profile of Cl-DGEBPA resin
confirms that the formation of DGEBPA resin pro- Effect of catalytic excess of sodium hydroxide load-
ceeds via formation of chlorohydrin intermediates. ing on the formation of DGEBPA resin and unde-
With the increase in polymerization temperature, for- sired byproducts was also studied at constant tem-
mation of chlorohydrins is also increased, and favors perature. Various amounts of NaOH [e.g., 2.5% (wt/
the formation of DGEBPA resin. wt), 5.0% (wt/wt), 10.0% (wt/wt), and 25.0% (wt/
The variation in molar concentration of a- wt)] were considered for polymerization keeping
DGEBPA resin with polymerization temperature is reaction temperature at 363 K. Details of the poly-
shown in Figure 3(e). The formation a-DGEBPA merization experiments are given in Table III. Varia-
resin steadily increases at lower temperature (i.e., tion of molar concentration of EP, BPA, DGEBPA,
333 K) and attains a steady value at the end of poly- Cl-DGEBPA, and a-DGEBPA with polymerization
merization. Attainment of steady concentration pro- time are shown in Figure 4(ae) respectively. It is
file in reaction kinetics is an indication of equilib- found that molar concentration of EP falls rapidly
rium followed by the equal rates of formation and with decrease in alkali loading [Fig. 4(a)]. Similarly,
disappearance of the component.30 Here, reaction molar concentration of BPA falls rapidly with the
between DGEBPA resin and water results a- decrease in alkali loading [Fig. 4(b)]. It is also found
DGEBPA resin [Table II; eq. (4)], and the steady that molar concentration of BPA becomes nonzero
molar concentrations of a-DGEBPA resin in the pol- above 10% (wt/wt) alkali loading after allowing suf-
ymerizing mixture with reaction time at lower tem- ficient polymerization time. The decrease in conver-
perature are the indication of mild reversible reaction sion of EP and BPA at higher alkali loading is an in-
between DGEBPA resin and a-glycol. It is interesting dication of the inhibition of EP and BPA dissociation
to note that the formation of a-DGEBPA resin attains at higher alkali concentration. Therefore, it is desira-
a maximum value at higher temperature and there- ble to keep little catalytic excess of alkali concentra-
after molar concentration starts declining at the end tion to maintain minimum molar concentration of
of polymerization. It is also noted that molar concen- EP and BPA in the polymerizing mixture.
trations of DGEBPA resin almost remain unchanged Molar concentration of DGEBPA resin increases
at the end of polymerization [Fig. 3(c)], and corre- with alkali loading and results are shown in Figure
sponding molar concentration of EP [Fig. 3(a)] and 3(c). It is noted that molar concentration of Cl-
a-DGEBPA resin [Figs. 3(e)] falls rapidly with poly- DGEBPA resin vanishes quickly at higher alkali
merization time. This suggests the possibility of loading (i.e., 25.0, 10.0, and 5.0% at 363 K) and
forming a-DGEBPA resin substituted product and results are shown in Figure 4(d). Here, the rapid
formed due to the combination of a-DGEBPA resin depletion of Cl-DGEBPA resin [mainly, chlorohydrin
and EP. Therefore, the decrease in molar concentra- intermediates: Table II, eq. (1)] favors the formation
tion of a-DGEBPA resin is mainly due to formation DGEBPA resin [Fig. 4(c)]. Though the conversion of
of a-DGEBPA resin substituted product with EP at EP and BPA reduces with the increase in alkali load-
higher temperature [Fig. 3(e)]. Maximum molar con- ing [Fig. 4(a,b)], but the interconversion of chlorohy-
centration of a-DGEBPA resin also increases with drins to DGEBPA resin favors at higher alkali load-
increase in temperature, which indicates the forma- ing [Fig. 4(d)]. Therefore, overall DGEBPA formation
tion of a-DGEBPA resin substituted product and is enhanced with increase in alkali loading keeping
favors at the higher temperature. It is also men- minimum molar concentration of chlorohydrins [Fig.
tioned that the inter conversion of DGEBPA resin to 4(c)]. With the increase in alkali loading, molar con-
a-DGEBPA resin is much more easier due to rapid centration of a-DGEBPA resin attains a maximum
epoxy ring opening as compared with the reversed value and thereafter it decreases to a minimum
reaction and favors the coupling between a- value [Fig. 4(e)]. As explained earlier, the formation
DGEBPA resin and EP. In other words, depletion of of a-DGEBPA resin substituted EP formation during
a-DGEBPA resin favors the formation of a-DGEBPA polymerization favors at higher alkali loading.
resin substituted EP rather than DGEBPA resin at Therefore, higher alkali loading helps to increase
the higher temperature. Therefore, the higher tem- the yield of DGEBPA resin keeping minimum con-
perature enhances the conversion of EP and centration of chlorohydrins intermediate, whereas

Journal of Applied Polymer Science DOI 10.1002/app


2904 SREENADHAN AND GURIA

Figure 4 Variation in molar concentration with time under varying alkali loading for (a) epichlorohydrin, (b) bisphenol-
A, (c) diglycidyl ether of bisphenol-A, (d) chlorine substituted diglycidyl ether of bisphenol-A (chlorohydrins), and (e)
diol substituted diglycidyl ether of bisphenol-A.

CONCLUSIONS
conversion of EP and BPA is reduced with enhanced
formation of a-DGEBPA resin with the increase in Polymerization of EP and BPA with 10 : 1M ratio is
alkali loading at a given temperature. carried out in presence of catalytic excess of alkali.

Journal of Applied Polymer Science DOI 10.1002/app


RAPID MONITORING OF BISPHENOL-A DIGLYCIDYL ETHER 2905

Formation of monomeric DGEBPA is monitored berger, C. G., Kroschwitz, J. I., Eds.; John Wiley & Sons: New
using FT-IR technique through rapid identification York, 1986; Vol. 6, p 322.
4. Rozenberg, B. A. Adv Polym Sci 1986, 75, 113.
and quantification of desired product (DGEBPA), 5. Muskopf, J. W.; McCollister, S. B. In Dithiocarbamic Acid to
unreacted monomers (EP and BPA), and byproducts Ethanol. Ullmanns Encyclopedia of Industrial Chemistry; Ger-
(Cl-DGEBPA resin and a-DGEBPA resin). Several hartz, W., Yamamoto, Y. S., Kaudy, L., Rounsaville, J. F.,
calibration plots (percent absorbance vs. molar con- Schulz, G., Eds.; John Wiley & Sons: New York, 1987; Vol. A9,
centration) of EP, BPA, DGEBPA, Cl-DGEBPA, and p 547.
6. Buser, H. Environ Sci Technol 1986, 20, 404.
a-DGEBPA are presented for the standard samples 7. Elvia, M.; Mungui-Lopez, H.; Soto-Valdez, H. J Agric Food
at their characteristic transmittance peaks. Effects of Chem 2001, 49, 3666.
polymerization temperature and catalytic excess of 8. Suerio, R. A.; Araujo, M.; Suarez, S.; Garrido, M. J Mutagen
alkali loading on the desired product, and byprod- 2001, 16, 303.
ucts are studied in an automatic temperature con- 9. Hemminki, K.; Paasivirta, J.; Kurkirinne, T.; Virkki, L. Chem
Biol Interact 1980, 30, 259.
trolled autoclave. Temperature and catalytic excess 10. Lapkin, M.; Weir, E. D. In Encyclopaedia of Industrial Chemi-
of alkali have a marked effect on DGEBPA resin for- cal Analysis; Snell, F. D., Ettre, L. S., Eds.; Interscience Pub-
mation, byproducts formation, and unreacted mono- lisher: New York, 1971; Vol. 12, p 232.
mers. Because of the presence of conflicting objec- 11. Stenmark, G. A.; Weiss, F. T. Anal Chem 1958, 30, 381.
tives i.e., maximizing the yield of DGEBPA resin 12. Batzer, H.; Zahir, S. A. J Appl Polym Sci 1977, 21, 1843.
13. Antal, I.; Fuzes, L.; Samay, S.; Csillag, L. J Appl Polym Sci
and minimizing the yield of byproducts, it is very 1981, 26, 2783.
difficult to select unique combination of the process 14. Jungang, G. J Appl Polym Sci 1993, 48, 237.
parameters (i.e., temperature and alkali loading) for 15. Medjitov, D. R.; Shode, L. G.; Tseitlin, G. M. Polym Bull 1998,
which both objectives will be satisfied. Low temper- 40, 509.
ature and low catalytic excess of alkali helps to 16. Kuptsov, A. H. Z.; Hizhin, G. N. Handbook of Fourier Trans-
form Raman and Infrared Spectra of Polymers; Elsevier: Am-
maintain low molar concentration of BPA in poly- sterdam, 1998.
merizing mixture. Conversion of EP is favored at 17. Rouessac, F.; Rouessac, A. Chemical Analysis Modern Instru-
higher temperature and lower catalytic excess of al- mentation Methods and Techniques; John Wiley & Sons: New
kali. Higher temperature and increased catalytic York, 2007.
excess of alkali favors the formation DGEBPA resin. 18. Dannenberg, H.; Harp, W. R. Anal Chem 1956, 28, 86.
19. Mertzel, E.; Koenig, J. L. Adv Polym Sci 1986, 75, 73.
High catalytic excess of alkali and low temperature 20. Shreve, O. D.; Heether, M. R.; Knight, H. B.; Swern, D. Anal
helps to keep Cl-DGEBPA resin concentration as low Chem 1951, 23, 277.
as possible in the polymerization mixture. Molar 21. Lakshmi, M. S.; Reddy, B. S. R. Eur Polym Mater 2002, 38,
concentration of a-DGEBPA resin attains a maxi- 795.
mum for all the polymerizing mixtures except at the 22. Chen, B. F. C.; Hang, W.; Shuwen, P.; Xiuyan, W.; Lisong, D.
Eur Polym Mater 1939 2003, 39.
lowest polymerization temperature studied for ep- 23. Badawi, H. M.; Forner, W. Spectrochim Acta Part A 2008, 71,
oxy polymerization. 1095.
24. Lee, M. J.; Hur, S. W.; Durig, J. R. J Mol Struct 1998, 444, 99.
The authors are indebted to the Department of Polymer Engi- 25. Perrin, D. D.; Armarego, W. I. F.; Perrin, D. R. Purification of
neering (Birla Institute of Technology, Mesra, Ranchi 835215, Laboratory Chemicals; Pergamon Press: Oxford, 1980.
India) for providing the research facilities. 26. Sorenson, W. R.; Sweeny, W.; Campbell, T. W. Preparative
Methods of Polymer Chemistry; John Wiley & Sons, Inc.: New
York, 2001.
References 27. Weaver, D. G.; Smart, J. L. Ind Eng Chem 1959, 51, 894.
28. Lintschinger, J.; Rauter, W. Eur Food Res Technol 2006, 211,
1. Prime, R. B. In Thermal Characterization of Polymeric Materi- 211.
als; Turi, E. A., Ed.; Academic Press: New York, 1981; p 435. 29. Morrison, R. T.; Boyd, R. N. Organic Chemistry, 6th ed.; Pren-
2. Paul, S. Surface Coatings, Science and Technology; John and tice Hall of India Private Limited: New Delhi, 2002; Chapters
Wiley & Sons: New York, 1985. 17 and 24.
3. McAdams, L. V.; Gannon, J. A. In Encyclopedia of Polymer 30. Levenspiel, O. Chemical Reaction Engineering, 3rd ed.; John
Science and Engineering; Mark, H. F., Bikales, N. M., Over- Wiley & Sons (Asia) Pte Ltd.: New Delhi, 2001; Chapter 8.

Journal of Applied Polymer Science DOI 10.1002/app

You might also like