- Email: [email protected]
Determination of epoxy groups in epoxy resins by reaction-based headspace gas chromatography
Accepted Manuscript Determination of epoxy groups in epoxy resins by reaction-based headspace gas chromatography Wei-Qi Xie, Xin-Sheng Chai PII:
S0142-9418(16)31385-X
DOI:
10.1016/j.polymertesting.2017.01.020
Reference:
POTE 4906
To appear in:
Polymer Testing
Received Date: 14 December 2016 Accepted Date: 22 January 2017
Please cite this article as: W.-Q. Xie, X.-S. Chai, Determination of epoxy groups in epoxy resins by reaction-based headspace gas chromatography, Polymer Testing (2017), doi: 10.1016/ j.polymertesting.2017.01.020. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Analysis Method Determination of Epoxy Groups in Epoxy Resins by Reaction-based Headspace Gas Chromatography
RI PT
Wei-Qi Xie1,2 and Xin-Sheng Chai1* 1. State Key Laboratory of Pulp and Paper Engineering, South China University of Technology, Guangzhou, China
2. School of Materials Science and Engineering, South China University of Technology,
SC
Guangzhou, China
M AN U
Abstract: This work reports on a new method for the determination of epoxy groups in epoxy resins by reaction-based headspace gas chromatography (HS-GC). After epoxy resins reacted with hydrochloric acid (HCl) solution, the remaining HCl reacted with bicarbonate solution in a closed headspace vial to form carbon dioxide that was
TE D
measured by HS-GC. It was found that the first reaction can be finished in 30 min at room temperature and the second reaction, together with headspace equilibration, can be achieved within 15 min at 60 oC. The results showed that the method has a good
EP
precision and accuracy, in which the relative standard deviation in the repeatability
AC C
measurement was 4.20%, and the relative differences between the data obtained by the HS-GC method and the reference method were within 8.04%. The present method is simple, efficient, and suitable for the used in the epoxy resin related research and applications.
Keywords: Epoxy groups; Headspace; Gas chromatography; Epoxy resins; Reaction-based 1
ACCEPTED MANUSCRIPT
1. Introduction Epoxy resins, the polymers containing epoxide groups [1], have been widely
RI PT
used in versatile applications, such as surface coatings, electrical laminates, adhesives and molding compounds [2-4]. Because the epoxy groups in polymers can significantly affect their performance in the application [5], analytical methods that
SC
can efficiently and accurately quantify the epoxy groups in epoxy resins are important
M AN U
to the process control for both the epoxy resin synthesis and product quality. Traditionally, epoxy group in epoxy resins is determined by the titration method [6]. In this method, the sample is dissolved in an organic (e.g., acetone or isopropanol) solution contains an excess amount of hydrochloric acid (HCl). After the reaction
TE D
between the epoxy group and HCl (to form hydroxyl), the unreacted HCl is titrated with potassium hydroxide, and thus the epoxy groups in epoxy resins can be determined. The major disadvantage of the method is the time-consuming procedures,
EP
which leads to a low efficiency in batch sample analysis.
AC C
There are also several methods available for the determination of epoxy group in epoxy resins based on advanced instruments, which include gas chromatography (GC) [7], Fourier transform infrared (FT-IR) spectroscopy [8-10] and 1H nuclear magnetic resonance (1H NMR) [11]. The conventional GC method is only suitable to be used for volatile compounds (e.g., monomers with low molecular weight), it failed to test non-volatile epoxy resins (polymers). Although FT-IR and NMR methods can conduct the measurement for both volatile and non-volatile compounds, these methods suffer 2
ACCEPTED MANUSCRIPT from technical difficulties, e.g., water content interference in FT-IR analyses [12] and incomplete solubility of the sample in a deuterated solvent used in NMR analyses. Headspace gas chromatography (HS-GC) is an effective tool for the quantitative
RI PT
analysis of volatile species in samples with complex matrices [13,14]. HS-GC is also able to determine some non-volatile species if they can be quantitatively converted to the related volatile species through some chemical reactions [15,16]. In a previous
SC
study [16], we reported a phase reaction conversion HS-GC technique for the
M AN U
determination of isocyanate groups in the organic intermediates. It was based on HS-GC measuring the carbon dioxide released from the reaction between the isocyanate groups and H2O in the solution. According to Payne et al. [17], the epoxy groups can react with excess hydrochloric acid to form hydroxyl. We believe that by
TE D
measuring the carbon dioxide generated from the unreacted HCl and bicarbonate, the epoxy groups in epoxy resins could be quantified. In this work, we proposed an automated HS-GC method for the determination of
EP
the epoxy groups in epoxy resins. The main focuses were to explore the reaction
AC C
conditions (i.e., the dosage of HCl and bicarbonate, sample size, reaction time and temperature) in order to achieve a complete epoxy groups conversion; and the operation conditions for headspace equilibration in the HS-GC measurement. The precision and accuracy of the present method in the epoxy group analysis were also evaluated.
2. Experimental 3
ACCEPTED MANUSCRIPT 2.1. Chemicals and Materials All chemicals used in this work (i.e., isopropanol, sodium bicarbonate and hydrochloric acid) were analytical grade purchased from commercial sources and
RI PT
used without further purification. A set of epoxy standard solutions was prepared by diluting the original epoxy with appropriate amounts of isopropanol and then calibrated by titration with hydrochloric acid standard solutions. The epoxy resin
SC
samples for the test were obtained from both manufacturers and a chemical supplier
2.2 Apparatus and operations
M AN U
(Aladdin Reagent Co. Ltd, China).
The HS-GC measurements were carried out with an automated headspace sampler (Thermo HS TriPlus 300, US) connected to a GC system (Agilent GC 7890A,
TE D
US) equipped with a thermal conductivity detector (TCD) and a GS-Q capillary column (length = 30 m, inner diameter = 0.32 mm with thickness of stationary phase = 1 µm), operating at a temperature of 105 °C with nitrogen carrier gas (flow rate =
EP
2.7 mL/ min). The headspace operating conditions were as follows: 15 min of strong
AC C
shaking to allow sample equilibration at 60 °C; sample loop temperature = 70 °C; transfer line temperature = 80 °C; pressurization pressure = 1.00 bar; carrier gas pressure = 1.50 bar; vial pressurization time = 15 s; sample loop fill time = 10 s; and transfer time = 20 s; sample loop volume = 3 mL. 2.3. Procedures for sample preparation and HS-GC measurement A 10.0 mg of epoxy resin sample was added to a 20 mL headspace sample vial that contained 1 mL overdose of HCl isopropanol solution. After the epoxy resin 4
ACCEPTED MANUSCRIPT sample had completely reacted with HCl, 4 mL of bicarbonate solution was added to the headspace sample vial. The vial was immediately sealed by a septum and then placed on the tray of the headspace sampler for automatic HS-GC testing. The
2.4. Determination of epoxy groups by a reference method
RI PT
reaction/equilibration was conducted at 60 °C for 15 min and measured by GC.
A given amount of epoxy resin was first dissolved with a certain amount of
SC
organic solvent (e.g., acetone or isopropanol) containing an excess amount of HCl.
M AN U
After the reaction between the epoxy group and HCl, the excess amount of HCl was back-titrated with a standard KOH solution, and then the epoxy groups in the epoxy samples can be calculated.
TE D
3. Results and discussion
3.1. Conditions for the ring-opening reaction Putting epoxy resin into HCl containing isopropanol solution results in the follow
HCl
OHCl
(1)
AC C
O
EP
reaction taking place
The epoxy groups in the resin sample can be completely converted to
chlorohydrins if HCl is overdosed. The unreacted HCl in the liquid medium can react with bicarbonate to form CO2, which can be easily measured by HS-GC. Because the epoxy resin (polymer) is insoluble in the aqueous solution, an organic solvent must be chosen. In the traditional method, acetone or isopropanol is used as solvent for such a purpose. However, since the headspace equilibration for quantifying CO2 is generally 5
ACCEPTED MANUSCRIPT operated at an elevated temperature (e.g., 60 °C), the use of acetone (boiling point = 58 °C) can cause a higher vapor pressure and thus reduce the detection sensitivity for CO2 in HS-GC analysis. Therefore, we chose isopropanol (boiling point = 82 °C) as
RI PT
the solvent in the present work. Fig. 1 shows the time-dependent CO2 formation from the epoxy reaction with HCl at the given conditions (magnetic stirring and room temperature). It can be seen that a complete conversion reaction of epoxy resin is
M AN U
SC
achieved in 30 min.
3.2. Conditions for the headspace analysis
In the present method, the content of the unreacted HCl in the above sample medium after reaction is indirectly determined by measuring the CO2 released from
TE D
the following reaction [14],
H + + HCO3− → H 2O + CO2 ( g )
(2)
There, the dosage of bicarbonate, the temperature and time for the CO2 formation in
EP
the above reaction and headspace equilibration are important to the quantification
AC C
analysis of the unreacted HCl. An excess amount of bicarbonate is required to ensure complete neutralization of the H+. However, if the concentration of bicarbonate added is too high, it could be decomposed into CO2, especially at an elevated temperature. The higher background in blank testing affects the detection limit of the method. Therefore, the dosage of bicarbonate added in the reaction system should be slightly above the unreacted HCl used in the testing. As shown in Fig. 2, it was found that the complete CO2 release from Reaction (2) 6
ACCEPTED MANUSCRIPT at the given medium can be achieved in 15 min at 60 °C. Higher temperatures could be helpful to speed up the reaction and the headspace equilibration, however it also has a risk of the bicarbonate decomposing [15]. Therefore, a mild temperature (60 oC)
RI PT
was chosen in the following investigation.
3.3. Chromatographic separation
SC
Because air was included in the headspace vial during the test, nitrogen, oxygen
M AN U
and water vapor may could affect the measurement of CO2. Since nitrogen was used as the GC carrier gas in the present method, there is no response of nitrogen signal detected by TCD in the GC system. As shown in Fig. 3, two species, i.e., oxygen and H2O can be well-separated from CO2 in the chromatogram at the given GC conditions.
TE D
Therefore, the CO2 can be accurately determined by the present method. It should be pointed out that the small amount of CO2 in air could affect the result when the amount of unreacted HCl is very low in the samples. However, such
AC C
EP
an effect can be simply subtracted by running a blank test.
3.4. Sample size
In the present method, the given HCl solution contains 1.25 mmol of hydrogen
ion (50 mL of 0.025 mol/L). Therefore, the epoxy group content in epoxy resin samples added in the above solution should be lower than 1.25 mmol according to Eq. (1). Thus, the maximum sample size used in the test is limited by the amount of reaction agent (i.e., the HCl solution). Fig. 4 shows the GC signal response to the 7
ACCEPTED MANUSCRIPT sample size (corresponding to the amount of epoxy groups) added in the reaction. The results indicate that there is a linear relationship between the GC signal for CO2 measurement and the size of samples until it is greater than 0.8 mmol at the given
RI PT
conditions.
3.5. Method evaluation
SC
3.5.1. Method calibration
M AN U
An external standard calibration was employed in the present HS-GC method. The calibration was performed by adding 1 mL of HCl standard solutions described in experimental section to a set of headspace sample vials that containing 4.0 mL NaHCO3 solution. These vials were then tested by HS-GC at the suggested conditions
TE D
mentioned above. A calibration equation was obtained which can be expressed as A = 0 .52 ( ± 0 .78 ) + 168 .5 ( ± 2 .24 ) × n remaining
or
(n = 7,
EP
A = a ( ±∆a ) + s ( ±∆s ) × nremaining
R2 = 0.999)
(3)
(3-1)
AC C
where A is the GC response to the CO2 released from the reaction and nr is the amount of remaining HCl (in mmol), a, ∆a and s represent the intercept, uncertainty of the intercept and slope in Eq. (3-1), respectively. The epoxy groups can be calculated by Eq. (4), i.e., A−a ntotal − s ntotal s − A + a C= = m ms
(4)
where nt, C and m represent the total amount of HCl (in mmol) added at the beginning, 8
ACCEPTED MANUSCRIPT the epoxy content of the sample (mmol/g) and the sample weight (g). The limit of quantitation (LOQ) in the present method was 5.34×10-2 mmol, which was calculated by the following equation. 10 ∆a s
(5)
RI PT
LOQ =
3.5.2. Method precision and validation
SC
The repeatability of the present method was evaluated by quadruplicate analysis of three epoxy resin samples from both industry and Aladdin Reagent Co. Ltd. The
were less than 4.20 % (see Table 1).
M AN U
results showed that the relative standard deviations (RSD) of these measurements
To validate the present method, a recovery experiment was conducted. We prepared a set of sample solutions by accurately spiking known amounts of epoxy into
TE D
epoxy resin samples. Thus, the net contribution from the added epoxy in the measurement for these spiked samples can be obtained. The results between the
EP
experimentally measured amounts by the present method and the spiked known amounts of epoxy actually added to the epoxy resin samples are shown in Table 2. It
AC C
can be seen that the recoveries of the present method are in the range of 97.9 – 102%, indicating that this new method has good accuracy in the epoxy quantification analysis.
The present method was also validated by comparing the results of the present HS-GC method with the results from a reference method (i.e., the back-titration method). As shown in Table 3, the epoxy groups of these epoxy resin samples measured by the present method have a good match with the results measured by the 9
ACCEPTED MANUSCRIPT reference method, in which the relative differences between these two methods are no more than 8.04%. Therefore, the present HS-GC method is suitable to be used for determining the epoxy groups of epoxy resin samples.
RI PT
3.5.3 Efficiency of the method Because commercial headspace auto-samplers have a functional mode called “overlapping thermo-stating”, the actually time for an individual vial measurement
SC
(including the second reaction) in the batch sample testing can be reduced to 5 -10
M AN U
min in the HS-GC measurement (including the second reaction/headspace equilibration) in the batch sample analysis.
For example, if we have 10 samples
need to be analyzed, the average time for a single sample in the present method is (45 + 10*3)/10 = 7.5 min (here we set 3 min as the interval testing time between two
TE D
samples).
The present method also saves a significant amount of chemicals when
test.
AC C
Conclusions
EP
comparing with the titration method, due to the smaller sample size, e.g., 0.2 g, in the
An HS-GC method for the quantification of epoxy groups in epoxy resins is
proposed.
Isopropanol was used as the solvent to dissolve epoxy resin sample and
then reacted with overdose HCl solution, and the unreacted HCl was further react with bicarbonate to CO2 in a closed vial following by HS-GC measurement. The new method is simple, accurate and efficient, it is particularly suitable for the batch analysis of epoxy groups in epoxy resins. 10
ACCEPTED MANUSCRIPT
Acknowledgements
RI PT
The authors acknowledge the financial support from the Key Project of the Science and Technology Ministry of Guangzhou, China (Grant No. 201607020025) and the National Natural Science Foundation of China (Project Nos. 21576105 and
M AN U
SC
51409287).
References
[1] K. Dušek, M. Bleha, S. Luňák, Curing of epoxide resins: Model reactions of curing with amines, J. Polym. Sci. Pol. Chem. 15 (1977) 2393-2400.
64 (1999) 427-431.
TE D
[2] S. Hörold, Phosphorus flame retardants in thermoset resins, Polym. Degrad. Stabil.
EP
[3] P. L. Teh, M. Mariatti, H. M. Akil, C. K. Yeoh, K. N. Seetharamu, A. N. R. Wagiman, K. S. Beh, The properties of epoxy resin coated silica fillers composites,
AC C
Mater. Lett. 61 (2007) 2156-2158. [4] H. Wang, W. Han, H. Tian, Y. Wang, The preparation and properties of glass powder reinforced epoxy resin, Mater. Lett. 59 (2005) 94-99. [5] H. Dannenberg, Determination of functional groups in epoxy resins by near‐ infrared spectroscopy, Polym. Eng. Sci. 3 (1963) 78-88. [6] H. Lee, K. Neville, Handbook of epoxy resins, McGraw-Hill, New York, 1967. [7] V. J. van Hylckama, W. De Koning, D. B. Janssen, Transformation kinetics of 11
ACCEPTED MANUSCRIPT chlorinated ethenes by Methylosinus trichosporium OB3b and detection of unstable epoxides by on-line gas chromatography, Appl. Environ. Microb. 62 (1996) 3304-3312.
RI PT
[8] L. Xu, J. H. Fu, J. R. Schlup, In situ near-infrared spectroscopic investigation of epoxy resin-aromatic amine cure mechanisms, J. Am. Chem. Soc. 116 (1994) 2821-2826.
SC
[9] R. F. Goddu, D. A. Delker, Determination of terminal epoxides by near-infrared
M AN U
spectrophotometry, Anal. Chem. 30 (1958) 2013-2016.
[10] M. Pramanik, S. K. Mendon, J. W. Rawlins, Determination of epoxy equivalent weight of glycidyl ether based epoxides via near infrared spectroscopy, Polym. Test. 31 (2012) 716-721.
TE D
[11] H. A. Aerts, P. A. Jacobs, Epoxide yield determination of oils and fatty acid methyl esters using 1H NMR, J. Am. Oil Chem. Soc. 81 (2004) 841-846. [12] T. Maruyama, S. Katoh, M. Nakajima, H. Nabetani, T. P. Abbott, A. Shono, K.
EP
Satoh, FT-IR analysis of BSA fouled on ultrafiltration and microfiltration membranes,
AC C
J. Membrane Sci. 192 (2001) 201-207. [13] B. Kolb, L. S. Ettre, Static Headspace-Gas Chromatography-Theory and Practice, 2nd ed., Wiley-VCH Press, New York, 2006. [14] W. Q. Xie, X. S. Chai, Rapid determination of moisture content in paper materials by multiple headspace extraction gas chromatography, J. Chromatogr. A 1443 (2016) 62-65. [15] W. Q. Xie, X. S. Chai, An efficient method for determining the acid value in 12
ACCEPTED MANUSCRIPT edible oils by solvent-assisted headspace gas chromatography, Anal. Methods 8 (2016) 5789-5793. [16] W. Q. Xie, X. S. Chai, Determination of isocyanate groups in the organic
RI PT
intermediates by reaction-based headspace gas chromatography, J. Chromatogr. A 1468 (2016) 241-244.
[17] G. B. Payne, P. H. DEMING, P. H. WILLIAMS, Reactions of hydrogen peroxide.
SC
VII. Alkali-catalyzed epoxidation and oxidation using a nitrile as co-reactant. J. Org.
AC C
EP
TE D
M AN U
Chem. 26 (1961) 659-663.
Figure Captions
Fig. 1 Effect of the time on the conversion of epoxy groups. (Note: GC signal is proportional to the amount of unreacted HCl) Fig. 2 Effect of time on the CO2 formation. Fig. 3 Chromatogram of a sample from HS-GC testing. 13
ACCEPTED MANUSCRIPT
M AN U
SC
RI PT
Fig. 4 The permitted amount of epoxy resin at the given conditions.
TE D
Tables
Table 1 The repeatability of the method epoxy content, mmol/g
Replica no.
Sample 2
Sample 3
0.89
1.10
1.82
0.86
1.16
1.80
3
0.81
1.19
1.87
4
0.86
1.13
1.89
Average
0.86
1.15
1.85
RSD, %
3.32
3.87
4.20
1
AC C
2
EP
Sample 1
14
ACCEPTED MANUSCRIPT Table 2 Recovery test Sample ID
Ethanol, mg/L
Recovery
Measured
1
1.45
1.48
102
2
0.94
0.92
97.9
3
1.15
1.17
4
1.31
1.29
5
1.86
1.84
RI PT
Added
102
98.5
M AN U
SC
99.0
Table 3 Comparison of the methods on epoxy group analysis Epoxy content, mmol/g Sample ID
2 3
AC C
4 5
Ref. method* (n = 3)
1.53±0.03
1.46±0.04
4.79
1.78±0.05
1.89±0.06
-5.82
1.21±0.02
1.12±0.04
8.04
0.78±0.03
0.81±0.03
-3.7
0.86±0.03
0.89±0.02
-3.4
EP
1
TE D
Present method (n = 3)
Relative difference, %
* The back-titration method
15
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
S0142-9418(16)31385-X
DOI:
10.1016/j.polymertesting.2017.01.020
Reference:
POTE 4906
To appear in:
Polymer Testing
Received Date: 14 December 2016 Accepted Date: 22 January 2017
Please cite this article as: W.-Q. Xie, X.-S. Chai, Determination of epoxy groups in epoxy resins by reaction-based headspace gas chromatography, Polymer Testing (2017), doi: 10.1016/ j.polymertesting.2017.01.020. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Analysis Method Determination of Epoxy Groups in Epoxy Resins by Reaction-based Headspace Gas Chromatography
RI PT
Wei-Qi Xie1,2 and Xin-Sheng Chai1* 1. State Key Laboratory of Pulp and Paper Engineering, South China University of Technology, Guangzhou, China
2. School of Materials Science and Engineering, South China University of Technology,
SC
Guangzhou, China
M AN U
Abstract: This work reports on a new method for the determination of epoxy groups in epoxy resins by reaction-based headspace gas chromatography (HS-GC). After epoxy resins reacted with hydrochloric acid (HCl) solution, the remaining HCl reacted with bicarbonate solution in a closed headspace vial to form carbon dioxide that was
TE D
measured by HS-GC. It was found that the first reaction can be finished in 30 min at room temperature and the second reaction, together with headspace equilibration, can be achieved within 15 min at 60 oC. The results showed that the method has a good
EP
precision and accuracy, in which the relative standard deviation in the repeatability
AC C
measurement was 4.20%, and the relative differences between the data obtained by the HS-GC method and the reference method were within 8.04%. The present method is simple, efficient, and suitable for the used in the epoxy resin related research and applications.
Keywords: Epoxy groups; Headspace; Gas chromatography; Epoxy resins; Reaction-based 1
ACCEPTED MANUSCRIPT
1. Introduction Epoxy resins, the polymers containing epoxide groups [1], have been widely
RI PT
used in versatile applications, such as surface coatings, electrical laminates, adhesives and molding compounds [2-4]. Because the epoxy groups in polymers can significantly affect their performance in the application [5], analytical methods that
SC
can efficiently and accurately quantify the epoxy groups in epoxy resins are important
M AN U
to the process control for both the epoxy resin synthesis and product quality. Traditionally, epoxy group in epoxy resins is determined by the titration method [6]. In this method, the sample is dissolved in an organic (e.g., acetone or isopropanol) solution contains an excess amount of hydrochloric acid (HCl). After the reaction
TE D
between the epoxy group and HCl (to form hydroxyl), the unreacted HCl is titrated with potassium hydroxide, and thus the epoxy groups in epoxy resins can be determined. The major disadvantage of the method is the time-consuming procedures,
EP
which leads to a low efficiency in batch sample analysis.
AC C
There are also several methods available for the determination of epoxy group in epoxy resins based on advanced instruments, which include gas chromatography (GC) [7], Fourier transform infrared (FT-IR) spectroscopy [8-10] and 1H nuclear magnetic resonance (1H NMR) [11]. The conventional GC method is only suitable to be used for volatile compounds (e.g., monomers with low molecular weight), it failed to test non-volatile epoxy resins (polymers). Although FT-IR and NMR methods can conduct the measurement for both volatile and non-volatile compounds, these methods suffer 2
ACCEPTED MANUSCRIPT from technical difficulties, e.g., water content interference in FT-IR analyses [12] and incomplete solubility of the sample in a deuterated solvent used in NMR analyses. Headspace gas chromatography (HS-GC) is an effective tool for the quantitative
RI PT
analysis of volatile species in samples with complex matrices [13,14]. HS-GC is also able to determine some non-volatile species if they can be quantitatively converted to the related volatile species through some chemical reactions [15,16]. In a previous
SC
study [16], we reported a phase reaction conversion HS-GC technique for the
M AN U
determination of isocyanate groups in the organic intermediates. It was based on HS-GC measuring the carbon dioxide released from the reaction between the isocyanate groups and H2O in the solution. According to Payne et al. [17], the epoxy groups can react with excess hydrochloric acid to form hydroxyl. We believe that by
TE D
measuring the carbon dioxide generated from the unreacted HCl and bicarbonate, the epoxy groups in epoxy resins could be quantified. In this work, we proposed an automated HS-GC method for the determination of
EP
the epoxy groups in epoxy resins. The main focuses were to explore the reaction
AC C
conditions (i.e., the dosage of HCl and bicarbonate, sample size, reaction time and temperature) in order to achieve a complete epoxy groups conversion; and the operation conditions for headspace equilibration in the HS-GC measurement. The precision and accuracy of the present method in the epoxy group analysis were also evaluated.
2. Experimental 3
ACCEPTED MANUSCRIPT 2.1. Chemicals and Materials All chemicals used in this work (i.e., isopropanol, sodium bicarbonate and hydrochloric acid) were analytical grade purchased from commercial sources and
RI PT
used without further purification. A set of epoxy standard solutions was prepared by diluting the original epoxy with appropriate amounts of isopropanol and then calibrated by titration with hydrochloric acid standard solutions. The epoxy resin
SC
samples for the test were obtained from both manufacturers and a chemical supplier
2.2 Apparatus and operations
M AN U
(Aladdin Reagent Co. Ltd, China).
The HS-GC measurements were carried out with an automated headspace sampler (Thermo HS TriPlus 300, US) connected to a GC system (Agilent GC 7890A,
TE D
US) equipped with a thermal conductivity detector (TCD) and a GS-Q capillary column (length = 30 m, inner diameter = 0.32 mm with thickness of stationary phase = 1 µm), operating at a temperature of 105 °C with nitrogen carrier gas (flow rate =
EP
2.7 mL/ min). The headspace operating conditions were as follows: 15 min of strong
AC C
shaking to allow sample equilibration at 60 °C; sample loop temperature = 70 °C; transfer line temperature = 80 °C; pressurization pressure = 1.00 bar; carrier gas pressure = 1.50 bar; vial pressurization time = 15 s; sample loop fill time = 10 s; and transfer time = 20 s; sample loop volume = 3 mL. 2.3. Procedures for sample preparation and HS-GC measurement A 10.0 mg of epoxy resin sample was added to a 20 mL headspace sample vial that contained 1 mL overdose of HCl isopropanol solution. After the epoxy resin 4
ACCEPTED MANUSCRIPT sample had completely reacted with HCl, 4 mL of bicarbonate solution was added to the headspace sample vial. The vial was immediately sealed by a septum and then placed on the tray of the headspace sampler for automatic HS-GC testing. The
2.4. Determination of epoxy groups by a reference method
RI PT
reaction/equilibration was conducted at 60 °C for 15 min and measured by GC.
A given amount of epoxy resin was first dissolved with a certain amount of
SC
organic solvent (e.g., acetone or isopropanol) containing an excess amount of HCl.
M AN U
After the reaction between the epoxy group and HCl, the excess amount of HCl was back-titrated with a standard KOH solution, and then the epoxy groups in the epoxy samples can be calculated.
TE D
3. Results and discussion
3.1. Conditions for the ring-opening reaction Putting epoxy resin into HCl containing isopropanol solution results in the follow
HCl
OHCl
(1)
AC C
O
EP
reaction taking place
The epoxy groups in the resin sample can be completely converted to
chlorohydrins if HCl is overdosed. The unreacted HCl in the liquid medium can react with bicarbonate to form CO2, which can be easily measured by HS-GC. Because the epoxy resin (polymer) is insoluble in the aqueous solution, an organic solvent must be chosen. In the traditional method, acetone or isopropanol is used as solvent for such a purpose. However, since the headspace equilibration for quantifying CO2 is generally 5
ACCEPTED MANUSCRIPT operated at an elevated temperature (e.g., 60 °C), the use of acetone (boiling point = 58 °C) can cause a higher vapor pressure and thus reduce the detection sensitivity for CO2 in HS-GC analysis. Therefore, we chose isopropanol (boiling point = 82 °C) as
RI PT
the solvent in the present work. Fig. 1 shows the time-dependent CO2 formation from the epoxy reaction with HCl at the given conditions (magnetic stirring and room temperature). It can be seen that a complete conversion reaction of epoxy resin is
M AN U
SC
achieved in 30 min.
3.2. Conditions for the headspace analysis
In the present method, the content of the unreacted HCl in the above sample medium after reaction is indirectly determined by measuring the CO2 released from
TE D
the following reaction [14],
H + + HCO3− → H 2O + CO2 ( g )
(2)
There, the dosage of bicarbonate, the temperature and time for the CO2 formation in
EP
the above reaction and headspace equilibration are important to the quantification
AC C
analysis of the unreacted HCl. An excess amount of bicarbonate is required to ensure complete neutralization of the H+. However, if the concentration of bicarbonate added is too high, it could be decomposed into CO2, especially at an elevated temperature. The higher background in blank testing affects the detection limit of the method. Therefore, the dosage of bicarbonate added in the reaction system should be slightly above the unreacted HCl used in the testing. As shown in Fig. 2, it was found that the complete CO2 release from Reaction (2) 6
ACCEPTED MANUSCRIPT at the given medium can be achieved in 15 min at 60 °C. Higher temperatures could be helpful to speed up the reaction and the headspace equilibration, however it also has a risk of the bicarbonate decomposing [15]. Therefore, a mild temperature (60 oC)
RI PT
was chosen in the following investigation.
3.3. Chromatographic separation
SC
Because air was included in the headspace vial during the test, nitrogen, oxygen
M AN U
and water vapor may could affect the measurement of CO2. Since nitrogen was used as the GC carrier gas in the present method, there is no response of nitrogen signal detected by TCD in the GC system. As shown in Fig. 3, two species, i.e., oxygen and H2O can be well-separated from CO2 in the chromatogram at the given GC conditions.
TE D
Therefore, the CO2 can be accurately determined by the present method. It should be pointed out that the small amount of CO2 in air could affect the result when the amount of unreacted HCl is very low in the samples. However, such
AC C
EP
an effect can be simply subtracted by running a blank test.
3.4. Sample size
In the present method, the given HCl solution contains 1.25 mmol of hydrogen
ion (50 mL of 0.025 mol/L). Therefore, the epoxy group content in epoxy resin samples added in the above solution should be lower than 1.25 mmol according to Eq. (1). Thus, the maximum sample size used in the test is limited by the amount of reaction agent (i.e., the HCl solution). Fig. 4 shows the GC signal response to the 7
ACCEPTED MANUSCRIPT sample size (corresponding to the amount of epoxy groups) added in the reaction. The results indicate that there is a linear relationship between the GC signal for CO2 measurement and the size of samples until it is greater than 0.8 mmol at the given
RI PT
conditions.
3.5. Method evaluation
SC
3.5.1. Method calibration
M AN U
An external standard calibration was employed in the present HS-GC method. The calibration was performed by adding 1 mL of HCl standard solutions described in experimental section to a set of headspace sample vials that containing 4.0 mL NaHCO3 solution. These vials were then tested by HS-GC at the suggested conditions
TE D
mentioned above. A calibration equation was obtained which can be expressed as A = 0 .52 ( ± 0 .78 ) + 168 .5 ( ± 2 .24 ) × n remaining
or
(n = 7,
EP
A = a ( ±∆a ) + s ( ±∆s ) × nremaining
R2 = 0.999)
(3)
(3-1)
AC C
where A is the GC response to the CO2 released from the reaction and nr is the amount of remaining HCl (in mmol), a, ∆a and s represent the intercept, uncertainty of the intercept and slope in Eq. (3-1), respectively. The epoxy groups can be calculated by Eq. (4), i.e., A−a ntotal − s ntotal s − A + a C= = m ms
(4)
where nt, C and m represent the total amount of HCl (in mmol) added at the beginning, 8
ACCEPTED MANUSCRIPT the epoxy content of the sample (mmol/g) and the sample weight (g). The limit of quantitation (LOQ) in the present method was 5.34×10-2 mmol, which was calculated by the following equation. 10 ∆a s
(5)
RI PT
LOQ =
3.5.2. Method precision and validation
SC
The repeatability of the present method was evaluated by quadruplicate analysis of three epoxy resin samples from both industry and Aladdin Reagent Co. Ltd. The
were less than 4.20 % (see Table 1).
M AN U
results showed that the relative standard deviations (RSD) of these measurements
To validate the present method, a recovery experiment was conducted. We prepared a set of sample solutions by accurately spiking known amounts of epoxy into
TE D
epoxy resin samples. Thus, the net contribution from the added epoxy in the measurement for these spiked samples can be obtained. The results between the
EP
experimentally measured amounts by the present method and the spiked known amounts of epoxy actually added to the epoxy resin samples are shown in Table 2. It
AC C
can be seen that the recoveries of the present method are in the range of 97.9 – 102%, indicating that this new method has good accuracy in the epoxy quantification analysis.
The present method was also validated by comparing the results of the present HS-GC method with the results from a reference method (i.e., the back-titration method). As shown in Table 3, the epoxy groups of these epoxy resin samples measured by the present method have a good match with the results measured by the 9
ACCEPTED MANUSCRIPT reference method, in which the relative differences between these two methods are no more than 8.04%. Therefore, the present HS-GC method is suitable to be used for determining the epoxy groups of epoxy resin samples.
RI PT
3.5.3 Efficiency of the method Because commercial headspace auto-samplers have a functional mode called “overlapping thermo-stating”, the actually time for an individual vial measurement
SC
(including the second reaction) in the batch sample testing can be reduced to 5 -10
M AN U
min in the HS-GC measurement (including the second reaction/headspace equilibration) in the batch sample analysis.
For example, if we have 10 samples
need to be analyzed, the average time for a single sample in the present method is (45 + 10*3)/10 = 7.5 min (here we set 3 min as the interval testing time between two
TE D
samples).
The present method also saves a significant amount of chemicals when
test.
AC C
Conclusions
EP
comparing with the titration method, due to the smaller sample size, e.g., 0.2 g, in the
An HS-GC method for the quantification of epoxy groups in epoxy resins is
proposed.
Isopropanol was used as the solvent to dissolve epoxy resin sample and
then reacted with overdose HCl solution, and the unreacted HCl was further react with bicarbonate to CO2 in a closed vial following by HS-GC measurement. The new method is simple, accurate and efficient, it is particularly suitable for the batch analysis of epoxy groups in epoxy resins. 10
ACCEPTED MANUSCRIPT
Acknowledgements
RI PT
The authors acknowledge the financial support from the Key Project of the Science and Technology Ministry of Guangzhou, China (Grant No. 201607020025) and the National Natural Science Foundation of China (Project Nos. 21576105 and
M AN U
SC
51409287).
References
[1] K. Dušek, M. Bleha, S. Luňák, Curing of epoxide resins: Model reactions of curing with amines, J. Polym. Sci. Pol. Chem. 15 (1977) 2393-2400.
64 (1999) 427-431.
TE D
[2] S. Hörold, Phosphorus flame retardants in thermoset resins, Polym. Degrad. Stabil.
EP
[3] P. L. Teh, M. Mariatti, H. M. Akil, C. K. Yeoh, K. N. Seetharamu, A. N. R. Wagiman, K. S. Beh, The properties of epoxy resin coated silica fillers composites,
AC C
Mater. Lett. 61 (2007) 2156-2158. [4] H. Wang, W. Han, H. Tian, Y. Wang, The preparation and properties of glass powder reinforced epoxy resin, Mater. Lett. 59 (2005) 94-99. [5] H. Dannenberg, Determination of functional groups in epoxy resins by near‐ infrared spectroscopy, Polym. Eng. Sci. 3 (1963) 78-88. [6] H. Lee, K. Neville, Handbook of epoxy resins, McGraw-Hill, New York, 1967. [7] V. J. van Hylckama, W. De Koning, D. B. Janssen, Transformation kinetics of 11
ACCEPTED MANUSCRIPT chlorinated ethenes by Methylosinus trichosporium OB3b and detection of unstable epoxides by on-line gas chromatography, Appl. Environ. Microb. 62 (1996) 3304-3312.
RI PT
[8] L. Xu, J. H. Fu, J. R. Schlup, In situ near-infrared spectroscopic investigation of epoxy resin-aromatic amine cure mechanisms, J. Am. Chem. Soc. 116 (1994) 2821-2826.
SC
[9] R. F. Goddu, D. A. Delker, Determination of terminal epoxides by near-infrared
M AN U
spectrophotometry, Anal. Chem. 30 (1958) 2013-2016.
[10] M. Pramanik, S. K. Mendon, J. W. Rawlins, Determination of epoxy equivalent weight of glycidyl ether based epoxides via near infrared spectroscopy, Polym. Test. 31 (2012) 716-721.
TE D
[11] H. A. Aerts, P. A. Jacobs, Epoxide yield determination of oils and fatty acid methyl esters using 1H NMR, J. Am. Oil Chem. Soc. 81 (2004) 841-846. [12] T. Maruyama, S. Katoh, M. Nakajima, H. Nabetani, T. P. Abbott, A. Shono, K.
EP
Satoh, FT-IR analysis of BSA fouled on ultrafiltration and microfiltration membranes,
AC C
J. Membrane Sci. 192 (2001) 201-207. [13] B. Kolb, L. S. Ettre, Static Headspace-Gas Chromatography-Theory and Practice, 2nd ed., Wiley-VCH Press, New York, 2006. [14] W. Q. Xie, X. S. Chai, Rapid determination of moisture content in paper materials by multiple headspace extraction gas chromatography, J. Chromatogr. A 1443 (2016) 62-65. [15] W. Q. Xie, X. S. Chai, An efficient method for determining the acid value in 12
ACCEPTED MANUSCRIPT edible oils by solvent-assisted headspace gas chromatography, Anal. Methods 8 (2016) 5789-5793. [16] W. Q. Xie, X. S. Chai, Determination of isocyanate groups in the organic
RI PT
intermediates by reaction-based headspace gas chromatography, J. Chromatogr. A 1468 (2016) 241-244.
[17] G. B. Payne, P. H. DEMING, P. H. WILLIAMS, Reactions of hydrogen peroxide.
SC
VII. Alkali-catalyzed epoxidation and oxidation using a nitrile as co-reactant. J. Org.
AC C
EP
TE D
M AN U
Chem. 26 (1961) 659-663.
Figure Captions
Fig. 1 Effect of the time on the conversion of epoxy groups. (Note: GC signal is proportional to the amount of unreacted HCl) Fig. 2 Effect of time on the CO2 formation. Fig. 3 Chromatogram of a sample from HS-GC testing. 13
ACCEPTED MANUSCRIPT
M AN U
SC
RI PT
Fig. 4 The permitted amount of epoxy resin at the given conditions.
TE D
Tables
Table 1 The repeatability of the method epoxy content, mmol/g
Replica no.
Sample 2
Sample 3
0.89
1.10
1.82
0.86
1.16
1.80
3
0.81
1.19
1.87
4
0.86
1.13
1.89
Average
0.86
1.15
1.85
RSD, %
3.32
3.87
4.20
1
AC C
2
EP
Sample 1
14
ACCEPTED MANUSCRIPT Table 2 Recovery test Sample ID
Ethanol, mg/L
Recovery
Measured
1
1.45
1.48
102
2
0.94
0.92
97.9
3
1.15
1.17
4
1.31
1.29
5
1.86
1.84
RI PT
Added
102
98.5
M AN U
SC
99.0
Table 3 Comparison of the methods on epoxy group analysis Epoxy content, mmol/g Sample ID
2 3
AC C
4 5
Ref. method* (n = 3)
1.53±0.03
1.46±0.04
4.79
1.78±0.05
1.89±0.06
-5.82
1.21±0.02
1.12±0.04
8.04
0.78±0.03
0.81±0.03
-3.7
0.86±0.03
0.89±0.02
-3.4
EP
1
TE D
Present method (n = 3)
Relative difference, %
* The back-titration method
15
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT