MSc Chemistry
Analytical SciencesMaster Thesis
Identification and quantification of polyamides using
microwave digestion followed by gas and liquid
chromatography mass spectrometry and
13C-Nuclear
Magnetic Resonance.
by
Chantal Brouwer
11277823
June 2019 – March 2020 48 EC Examiners: Supervisor:Dhr. Prof. Dr. Ir. P.J. Schoenmakers Drs. N.J. van Beelen Dhr. Dr. A Gargano
Abstract
Polyamides occur naturally in silk and wool or can be made synthetic. Polyamides are polymers with amide linkages which hold the repeating units together. Polyamides are high-quality materials containing amide units which exhibit excellent thermal stability and physical properties. They are often synthesized from a diacid and diamine. A polyamide-modified alkyd resin is a thixotropic alkyd. During the polyamide modification step, the thixotropic alkyd is built into the alkyd resin. The thixotropic behavior of the polyamide-modified resin is determined by the hydrogen bonding between ester carbonyl and amide groups. The aim of this study was to identify and quantify polyamides (hardeners and amide modified alkyd resins) using the microwave digestion followed by GC-MS for amine, small acid and polyol identification and DIMS for dimeric acid. The digested polymer mixture(s) were also tested by NMR directly on monomer composition. A method from literature was tested using a reagent of 4M HCl in IPA during digestion and analyzed by GC-MS and NMR. The samples after the microwave digestion were never homogeneous and the two layers needed to be separated before analysis by GC-MS and NMR. The bottom layer consisted of amine and the top layer of fatty acids. In the bottom layer of the polyamide samples, 1,6-hexanediamine and 1,2-ethanediamine were identified. For the polyamide-modified alkyd resin, 2,4-toluenediamine was identified in the bottom layer. Quantification of polyamides samples was difficult. Very low recoveries were obtained by GC-MS (11-39%) and NMR (1.96-3.16%) using different extraction methods. After a titration of the fatty acids, the total conversion of all CRG samples was around 8-33%. To conclude, the most promising method was further tested using a reagent of 4M HCl in IPA during digestion. With this method it was possible to identify the amines for polyamides and polyamide-modified alkyd resins by GC-MS and NMR. Quantification still needs to be further optimized.
Abbreviations
COSY Correlation Spectroscopy
DEPT Distortedness Enhancement by Polarization Transfer
DIMS Direct Infusion Mass Spectrometry
DOSY Diffusion Ordered Spectroscopy
ERETIC Electronic REference To access In vivo Concentrations
GC-MS Gas chromatography – mass spectrometry
HMBC Heteronuclear Multiple Bond Correlation
HSQC Heteronuclear Single Quantum Coherence
IR Infrared
LC-MS Liquid chromatography – mass spectrometry
NMR Nuclear Magnetic Resonance
PA Phthalic anhydride
PT Polyol pentaerythritol
TMA Trimellitic anhydride
TMAH Tetramethylammonium hydroxide
TOCSY Total Correlation Spectroscopy
VOCs Volatile Organic Compounds
1. Introduction...1
1.1 Introduction...1
1.1.1 Alkyd resin...1
1.1.2 Polyamides...2
1.1.3 Polyamide-modified alkyd resin...2
1.1.4 Purpose of research...3 1.1.5 Aim of research...3 1.2 Techniques...4 1.2.1 Digestion methods...4 1.2.2 Microwave radiation...4 1.2.3 GC-MS...5 1.2.4 DIMS...6 1.2.5 NMR...6
1.3 Earlier research of polyamides...7
1.4 Sample information...7
1.5 Hypotheses...9
2. Alkyd resin identification...10
2.1 Chemicals...10
2.2 Methods...10
2.2.1 Microwave digestion method based on NMR...10
2.2.2 Microwave digestion method based on GC-MS...10
2.3 Results...10
2.3.1 NMR-analysis...11
2.3.2 GC-MS analysis...12
2.4 Conclusion...12
3. Polyamide identification and quantification...13
3.1 Test method exploration...13
3.2 Polyamide identification...15 3.2.1 Chemicals...15 3.2.2 GC-MS method...15 3.2.3 NMR method...15 3.2.4 Identification by GC-MS...16 3.2.5 Identification by NMR...21
3.2.6 Identification of dimeric fatty acids...24
3.3 Polyamide quantification...27
3.4 Summary polyamide identification and quantification...31
4. Polyamide-modified alkyd resin identification...32
5. Conclusion...35
6. Recommendations...35
8. Appendix...37
8.1 Alkyd resin identification...37
8.2 Standard solvent GC-MS settings...38
8.3 Column digestion in the oven...39
8.4 Polyamide identification by GC-MS...40
8.5 Polyamide identification by NMR...40
8.6 Polyamide identification by DIMS...46
1. Introduction
AkzoNobel is an internationally renowned paint and coatings company and produces paints, lacquers, coatings and special chemicals for consumers and industry worldwide. The well-known names such as, Flexa, Sikkens and Hammerite belong to AkzoNobel. AkzoNobel strives for a sustainable technology and innovative products and has a leadership position in the field of sustainability.
Paint is composed of a set of raw materials. The most important raw material is the binder, often referred as a resin, which ensures the formation of the dry paint film. Binders are solid or viscous liquid substances, which are usually used in combination with a solvent. The second most important raw material is the solvent. The solvent makes the paint processable in various steps of the production process. Organic solvents were often used, but nowadays many resins are dispersed in water to reduce the amount of volatile organic compounds (VOCs). It is important to switch from solvent-borne paints, which are harmful to the atmosphere and health, to waterborne paints. The third most important raw material is the pigment, the powdered color carrier. Finally, additives are added to improve properties of the paint. This combination together forms the liquid paint, which is converted into the dry paint film via various drying mechanisms. 1
1.1 Introduction
1.1.1 Alkyd resin
An alkyd is an oil-modified polyester by the addition of fatty acids. This resin is a polymer obtained from a polycondensation reaction of polyols, (such as pentaerythritol or propylene glycol), and polybasic acids (such as phthalic acid) or anhydride (e.g. phthalic anhydride). A simplified structure of alkyd resin made from glycerol, phthalic acid anhydride and linoleic acid is shown in figure 1.
Figure 1: The structure of an alkyd resin made from glycerol, phthalic acid anhydride and linoleic acid.
The variations in types of components give enormous varieties of resins with different properties. Besides to the actual alkyd resin vegetable oils are added, such as sunflower oil, safflower oil, or soybean oil. These low
vegetable oils ensure that the paints dry at room temperature. Drying time depends on the type and amount of oil that is used. Hereby, organic metal salts are mostly used as catalysts. 2
1.1.2 Polyamides
Polyamides occur naturally in silk and wool or can be made synthetic. Polyamides are polymers with amide linkages which hold the repeating units together. For example, a synthetic polyamide, nylon 66, is made from to monomers; hexanedioic acid and 1,6-hexanediamine and is shown in figure 2.
Figure 2: The reaction mechanism of polyamides synthesis, using two monomers; hexanedioic acid and 1,6-hexanediamine to form Nylon 66.
Polyamides are high-quality materials containing amide units which exhibit excellent thermal stability and physical properties. 3 They are often synthesized from a diacid and diamine. The polyamides, Nylon 66 and nylon 6, were invented in 1930s and are now around 90% of the total polyamide use and 99% of the polyamide fibers. 4 They can occur aliphatic, semi-aromatic or fully aromatic. High enough molecular weight aliphatic polyamide usable for industrial purposes are generally difficult to prepare. One of the most important difficulties is that very high temperatures and critical conditions are required during the polymerization process. 3
1.1.3 Polyamide-modified alkyd resin
A polyamide-modified alkyd resin is a thixotropic alkyd. During the polyamide modification step, the thixotropic alkyd is built into the alkyd resin. The thixotropic behavior of the polyamide-modified resin is determined by the hydrogen bonding between ester carbonyl and amide groups. Polyamide-modified alkyd resins prevent sagging so that higher layer thickness can be applied of the paint. The reduction of liquid viscosity introduced by shear forces is described by the term ‘thixotropy’. A thixotropic alkyd resin is an alkyd resin which is modified with a polyamide, polyaramid, or polyuria. By modifying the alkyd resin, hydrogen bonds are introduced into the binder. These can be broken down by shear stress, such as rolling or brushing due to the weak rheological structures. The viscosity substantially increases again after the solvent evaporates. 5, 6 The properties of structured viscous and thixotropic paints are good workability, better to apply in thicker layers and no sagging due
Polyamide-modified alkyd resin is synthesized by using polyamides. In general, the principle is based on the mix of polyamides containing thixotropic agents and alkyd resins during their polycondensation. To prepare a thixotropic binder by reacting a polyamide with an alkyd resin to obtain a modified alkyd resin needs at least dimeric fatty acid, diamine and monomeric acid. The polyamides are formed by amide linkages (see figure 2) and the monomeric acid can stop the polymerization. The diamines are used to form the thixotropic alkyd and dimer acids (C36) are obtained by the polymerization of unsaturated C18 fatty acids derived from monomers, dimers and trimers, such as oleic and linoleic acids. Many polyamides can be prepared by melt-polycondensation reactions using the dimeric acid and diamines. 7, 8, 9
1.1.4 Purpose of research
A purpose of paint test is rheological effects, better understanding of viscoelastic properties and to measure the amide alkyd and the possibility to change the amide alkyd for e.g. better drying, molecules better bonded, no sagging. Polyamide-modified alkyd resins prevent sagging so that higher layer thickness ca be applied of the paint. The amide composition influences the properties by H-bridging with each other or with the ester groups. One of the reasons to investigate polyamides is for competitor analysis. Another reason is to gain better control and analyze the composition of the resins. Hereby, is it possible to control the production resins on the correct composition and it can also be used to check whether restricted components do not exceed the maximum allowable quantity. Also, a better understanding between paint properties and the component in a specific resin.
1.1.5 Aim of research
The aim of this study is to identify and quantify polyamides (hardeners and amide modified alkyd resins) using the microwave digestion followed by gas chromatography mass spectrometry (GC-MS) for amine, small acid and polyol identification and liquid chromatography mass spectrometry (LC-MS) for dimeric acid. The digested polymer mixture(s) will also be tested by nuclear magnetic resonance (NMR) directly on monomer composition.
Microwave digestion can breakdown most polymer bonds in short analysis time. Currently, no digestion test method is available for polyamides. For this research an easy and safe digestion method had to be developed testing various acids and base digestions at various temperatures and duration. The digested mixture can be tested directly by 13C-NMR, GC-MS, and LCMS. The developed procedure should be verified by proprietary made polyamides from polymer laboratory and commercially available polyamide alkyd resins.
1.2 Techniques
1.2.1 Digestion methods
Digestion methods can be used to degrade polymers into monomers. 10 Identification and quantification of the monomer composition analysis of polymers could be simplified by a digestion method. Heating and/or radiating the sample can break specific bonds. When sufficient heat is applied, the bond energy is reached and breaks. Mostly, the weaker bonds will break in the polymer. When using radiation, ionic and dipolar bonds will break. In this research alkyd resin, polyamides, and polyamide-modified resins, which are containing dipolar bonds will be studied. Therefore, digestion methods based on radiation and/or heat can be useful to break polymers into monomers. 10, 11
1.2.2 Microwave radiation
A microwave uses electromagnetic energy and has a frequency range of 300-300,000 MHz. Hereby, only molecular rotation is affected. The microwave energy consists of a magnetic and electric field. Only the electric field can heat a substance by a transfer of energy. 12
Microwave heating is a non-conventional energy source which is often used in the organic chemistry. Due to the heating rate, magnificent accelerations can be achieved in a many different reactions, which is not possible with conventional heating. There are many advantages to microwave heating over conventional heating, such as rapid heating with high specificity, shorter reactions times, and milder reaction conditions. This can improve many reaction processes. 11 In table 1 the characteristics of the microwave and conventional heating are given. 13
Another main advantage is that the radiation penetrates in the bulk of the material and simultaneously heats the bulk of the material, see figure 3. 13 Microwaves can transfer energy directly to the reactive species, also called ‘molecular heating’, which can promote transformations that are currently not possible using conventional heating. The energy transmission is produced by dielectric losses, and the magnitude of heating is depending on these dielectric properties. Therefore, the absorption of radiation can be carried out selectively. However, in conventional heating, conduction and convection processes are observed. Microwave radiation is fast and volumetric, with the entire material being heated simultaneously. Conventional heating, on the other hand, is slow with heat introduced from the surface. 11, 13
Table 1: Characteristics of a microwave and conventional heating. 13
Microwave heating Conventional heating
Energetic coupling Conduction/convection Coupling at the molecular level Superficial heating
Rapid Slow
Volumetric Superficial
Selective Nonselective
material
Figure 3: The differences of heating using microwave radiation and conventional heating. On the left is a reagent tube using microwave radiation and on the right conventional heating is used. After 60 seconds the whole reaction volume is simultaneously heated on the left, in contrast with the conventional heating, where the surface and wall are heated first. 13 Most reactions are in solution, where the solvents play an important role. The choice of solvent can be crucial for the outcome of the reaction. Polarity is one of the most important characteristics of a solvent. Especially when using the microwave, which is coupling or in this case breaking the molecules directly in the reaction mixture. When the reaction mixture is more polar, the greater its ability to couple with the microwave energy. 12
In this research a CEM Discover SP-DTM microwave will be used, where SP-D stands for Sequential Pressurized SP-Digestion. In this microwave, each reaction mixture will be individually heated and destructed. Therefore, the temperature and pressure are controlled in each vessel.
1.2.3 GC-MS
GC-MS is commonly used method which combines the gas chromatography and mass spectrometer to identify and quantify volatile samples. In this research project GC-MS will be mostly used to identify and quantify the amine and small acids of the polyamides. For this analysis an Agilent 7890A gas chromatograph was used. The sample was separated on the GC with a non-polar stationary phase column CP9151 from Agilent (dimensions 0.25 m x 30 m x 0.25 mm) and analyzed with the MS Agilent 5975C inert XL EI/CI MSD with triple axis detector. The carrier gas is helium.
of non-volatile salt/ buffers and reagents. A disadvantage is that the mass accuracy can be adversely reduced when too many ions are trapped at one time. Therefore, DIMS is only suitable for pure or simple mixtures. In this research project the presence of dimeric fatty acids will be identified by direct infusion MS (DIMS). For the analysis a Bruker MicroTOF-Q mass spectrometer will be used. 14
1.2.5 NMR
NMR is besides mass spectrometry a very strong technique for the identification and quantification of organic molecules. More sample is needed for NMR than for MS. NMR is a non-destructive technique and it is expected that the complete analysis and interpretation are obtained of the entire spectrum. In NMR it is possible to measure in one dimensional 1H and 13C spectra or two dimensional. In two-dimensional NMR interactions and couplings can be studied between the protons and carbons (DEPT, HSQC, and HMBC) but also among protons (COSY and TOCSY) to give more inside information. 14
In Distortedness Enhancement by Polarization Transfer (DEPT) NMR can identify the number of protons coupled to a carbon, for example CH, CH2, CH3, or none. Heteronuclear Single Quantum Coherence (HSQC) can identify the correlations between the coupling of two nuclei, for example 1H spectrum on the x-axis and a 13C spectrum on the y-axis over 1 bond. The spectrum contains a spot for each unique 1H attached to the 13C being considered. In Heteronuclear Multiple Bond Correlation (HMBC) protons which are near but not directly bonded to different carbons can be studied. Combining HSQC with HMBC can give a lot of information on the molecular structure of the molecule. Besides the coupling correlation of 1H-13C, it is also possible to observe couplings between 13C-13C and 1H-1H. The 1H-1H coupling can be studies by Correlation Spectroscopy (COSY) and it give information about pairs of protons that are coupled. The protons are on adjacent carbons, for example 4 bonds away. In addition, Total Correlation Spectroscopy (TOCSY) give correlations for all protons within a spin system and the coupling may succeed over 5 bonds. Lastly, 1H of 13C Diffusion Ordered Spectroscopy (DOSY) NMR is also a two-dimensional NMR technique. It measures the proton or carbons during diffusions of the molecules in the z-axis. This spectrum will separate on the difference in size and proton shifts of the molecule. It is based on diffusion; smalls molecules move fast, and large molecules moves slow. In this case it would be possible to separate dimeric fatty acid from fatty acids and monomers. This technique can help the understanding of complex mixtures without the need of prior separation. It can also show if the digestion broke down dimer fatty acids (C36) into fatty acids (C18). 14
For the quantification of polyamides using NMR, an internal standard can be used. The internal standard must satisfy various conditions. The internal standard must be soluble in the sample. Also, an internal standard must not overlap or have a chemical interaction with the sample. Therefore, finding the ideal internal standard can be challenging. An alternative is the use of an ERETIC method (Electronic REference To access
In vivo Concentrations), which can determine absolute concentrations. In comparison to the internal standard, nothing is added to the sample and can be directly measured with the ERETIC method. Thus, all the disadvantages of the above-mentioned internal standard method are avoided. 15, 16
In this research project, H-NMR and 13C-NMR spectrum will be used for the identification of the amine, fatty acids, and dimeric fatty acids. 2D-NMR spectrum will be used to confirm the identification and to see correlations and bonds between the peaks and molecules. For the analysis a Bruker Avance III NMR spectrometer 400 MHz Ultrashield will be used.
1.3 Earlier research of polyamides
An earlier student here at AkzoNobel did research into digestion methods for polyamides using pyrolysis GC-MS and micro-reaction GC-MS. In principle, high temperature and pressure were used to break the polyamides and obtain amine and acid. The focus in the project was more on the acids. Dimeric fatty acids were analyzed and identified by LC-MS, but quantification needs still to be optimized. For the amines, commonly used diamines (1,2-ethanediamine and 1,6-hexanediamine) for polyamide synthesis were analyzed after thermochemolysis with tetramethylammonium hydroxide using micro-reaction GC-MS. Only, 1,6-hexanediamine could be identified after thermochemolysis GC-MS. However, when using a MicroJet Cryotrap before the GC column, identification 1,2-ethanediamine was possible. 17
A disadvantage was that only very small volumes could be measured and therefore analysis by NMR was not possible. In addition, this technique had a low sample through put and was very laborious. Approximate ten micro-reaction capillaries had to be prepared before it could be analyzed by LC-MS. Therefore, it was recommended to try another analysis approach. This method will be further explored. 17
1.4 Sample information
During this research, different samples are tested to gain knowledge in the behavior of amine. An alkyd resin, dimeric fatty acid, polyamides and unknown polyamides, and unknown polyamide-modified alkyd resin were investigated.
A commercially available alkyd resin was used to perform a digestion method by microwave, which was setup by an earlier student here at AkzoNobel. The monomer composition was polyol pentaerythritol, dibasic acids: trimellitic anhydride and phthalic anhydride, and fatty acids from sunflower oil. The structure formula of the monomers is shown in table 2.
Table 2: The monomer composition of commercially available alkyd resin with the structure formula of each monomer.
Monomer Percentage (%) Structure formula
Polyol pentaerythritol (PT) 17.6 OH OH O H O H
Trimellitic anhydride (TMA) 3.9
O H
O O
O
O
Phthalic anhydride (PA) 12.6
O O O
Fatty acids from sunflower oil
65.9
As a reference material a commercially available dimer fatty acid was used. Dimeric fatty acids are produced by dimerization of unsaturated C18 fatty acids, resulting in mixtures of C36 dimerized fatty acids and are composed of 0.1 %(wt) monomer, 1.0 %(wt) intermediate, 98 %(wt) dimer acid, and 1.0 %(wt) trimer. Mn = 570 g/mol.18
Four polyamides, referred as CRG 1-4, were obtained from the polymer laboratory of AkzoNobel Sassenheim and the monomer composition is shown in table 3. Unknown polyamides, referred as commercially available polyamides 1, 2, and 3 were obtained from Arizona Chemicals. One unknown commercially available polyamide-modified alkyd resin was also obtained from polymer laboratory of AkzoNobel Sassenheim.
Table 3: Monomer composition of four polyamides obtained from the polymer laboratory from AkzoNobel Sassenheim.
CRG1 CRG2 CRG3 CRG4
mass %
Dimeric fatty acid 88.75 69.10 66.92 56.58
1,6-hexanediamine 11.25 8.76 8.48 16.85
Isononanic acid 22.14
Sunflower fatty acid (HE30)
24.60
Tall oil fatty acid 26.57
1.5 Hypotheses
Here are the expected results which should be obtained when the amide is broken down from the polyamides and polyamide-modified alkyd resin. Three different techniques: NMR, GC-MS, and IR will be used for the identification of the amine. When the digestion is succeeded, amine and acids will be observed in all three methods.
The peaks expected in 13C-NMR spectrum after a successful digestion with the microwave are shown in table 4. Also, by comparing other NMR spectra from a commercially available alkyd resin and dimeric fatty acid spectra, which do not have an amide group, it would be easier to see any changes and extra peaks. In proton-NMR the amine peak will shift due to different environments and is really depending on the pH.
Table 4: Expected peaks in a 13C-NMR spectrum of amides and amines.
Functional group Structure Chemical shift (ppm)
Amine C-NR2 30-60
Amide R-CONR2 150-180
Carboxylic acid R-CO2H 160-190
Methanol CH3OH 49-50
TMAH C4H13NO 50
DMSO C₂H₆SO 38-40
IPA C3H7OH 25, 64
The GC-MS spectrum will be compared with the database to see if there are any amine fragments observed. Amines are expected around 1-4 minutes with this GC program. The peaks expected in an IR spectrum to observe either primary, secondary of tertiary amides are shown in table 5. 19
Table 5: Expected peaks in an IR spectrum of amides. 19
Primary amides/amines Secondary
amides/amines
Tertiary amides/amines
C-N stretch 1450 cm-1
Amide I band: C=O stretching 1681 cm-1
Amide band I: C=O stretching 1638 cm-1
Amide band I: C=O stretching 1640 cm-1
Amide band II: N-H bending in plane 1613
cm-1
Amide band II: N-H bending
in plane 1554 cm-1 Amide band II: C-N bendingin plane 1428 cm-1
Amine: N-H stretch, 2 bands
3149 cm-1 and 3325 cm-1
Amine: N-H stretch, 1 band
3326 cm-1 stretching bands aroundAmine: None N-H
3300 cm-1
Functional group Region (cm-1)
-CH2NH3+ / -NH3+ 1615-1560, 1635-1585, 3350-3100,
NH2+ (secondary amine) 1620-1560, 3000-2700
NH+ (tertiary amine) 2200-1800, 2700-2250
-COOH (saturated aliphatic
carboxylic acids) 1740-1700
-COOH (carboxylic acids) 1380-1280 -OH (associated carboxylic
2. Alkyd resin identification
2.1 Chemicals
The following chemicals were used for sample preparation of the alkyd: A commercially available alkyd resin, obtained from polymer lab at AkzoNobel Sassenheim, Tetramethylammonium hydroxide (TMAH) 25 wt% in methanol obtained from Alfa AesarTM Thermo Fisher (CAS: 75-59-2), methanol-d4 obtained from Across (99,8%) (CAS: 811-98-3), and methanol (HPLC grade) obtained from Baker (CAS: 67-56-1).
2.2 Methods
0.2 g of commercially available alkyd resin was dissolved in 0.5 mL TMAH (25 wt% in methanol) in a vessel and placed in the microwave Discover SP-DTM clinical (CEM). The digestion method consists a ramp and hold time of both 5 minutes at 120 °C. The speed of stirring during digestion was set on high.
2.2.1 Microwave digestion method based on NMR
For the method based on NMR, after digestion the solution was mixed with 0.3 mL methanol-d4 and added to an NMR tube. For the analysis a Bruker Avance III NMR spectrometer 400 MHz Ultrashield was used. A proton NMR (16 scans), 13C-NMR (2048 scans), 13C-DEPT-135 (256 scans) spectra were included.
2.2.2 Microwave digestion method based on GC-MS
For the method based on GC-MS, after digestion 0.3 mL methanol was added and the solution was placed in a GC vial. For this analysis an Agilent 7890A gas chromatograph was used. The sample was separated on the GC with a non-polar stationary phase column CP9151 from Agilent (dimensions 0.25 m x 30 m x 0.25 mm) and analyzed with the MS Agilent 5975C inert XL EI/CI MSD with triple axis detector. The inlet source was 280 °C and the inlet pressure was set to 55.8 kPa with a flowrate of 1 mL/min and a split ratio of 1:25. The carrier gas was helium. The GC temperature program is shown in table 6.
Table 6: GC temperature program for a commercially available alkyd resin.
Ramp (°C/min) Temperature
(°C) 0 50 20 107.5 5 150 15 185 5 200 15 238
2.3 Results
The prepared samples were measured by NMR and GC-MS for identification of the alkyd resin.
2.3.1 NMR-analysis
The samples were prepared and analyzed as described above. The aim of these experiments was to determine chemical shifts of certain components. After the analysis with the NMR, the monomers; polyol pentaerythritol (PT), the dibasic acids trimellitic anhydride (TMA) and phthalic anhydride (PA), and fatty acids were obtained. The total spectrum in shown in Appendix 8.1 figure 23. A clearer and more zoomed spectra is shown in figure 4 and figure 5.
1
2
6
5
4
3
9(CD
6 63
OD)
8 (PT)
7
10 (TMAH)
12 (PT)
11
O H
O H
OH
OH
8
12
12
12 12
C H
3
C H
3
C
H
3
N
+
C H
3
OH
-10
10
10
10
1,2,3,4,5,6 =
saturated
fatty acid
pattern
(MeOH)
Figure 4: A zoomed in 13C-NMR spectrum of commercially available alkyd resin from 0 to 70 ppm. Each peak represents a carbon of a monomer or an identified peak from (dimer) fatty acids. Peak 8 and 12 is the monomer polyol pentaerythritol (PT), peak 9 is methanol-d4, peak 11 is methanol, and peak 10 is tetramethylammonium hydroxide (TMAH).
14 13 17 (PA) (TMA) 19 20 21 22 22 23 24 Carboxyl group of dimer fatty acid 15 16 Carboxyl groups of fatty acid (TMA) 18 23 carboxyl group TMA
Figure 5: A zoomed in 13C-NMR spectrum of commercially available alkyd resin from 125 to 190 ppm. Each peak represents a carbon of a monomer or an identified peak from (dimer) fatty acids. Peak 17 is the monomer phthalic anhydride and peak 18 trimellitic acid.
The NMR spectra showed that it was possible to identify the monomers of the alkyd resin after microwave digestion at 120 °C with a ramp and hold time of 5 minutes. Also, different temperature (T140 or T160 °C) and hold times (H10 or H20 minutes) were tested to improve the method. However,
the NMR spectra did not change significantly. The method reached a maximum pressure of 300 psi at 160 °C. With longer hold times, the NMR spectra did not change either, so it was not necessary to extend the hold times.
2.3.2 GC-MS analysis
For analysis by GC-MS another sample was prepared and measured with the method described above. Also, a standard solvent GC temperature program from AkzoNobel was tested with a temperature ramp of 20 °C/min from 50 to 300 °C. The reason for this was to see the difference and optimization which was already done. Both chromatograms were compared with the database of AkzoNobel and Wiley/NIST, and all the monomer peaks of commercially available alkyd resin were identified in both chromatograms. However, the peak shapes with the standard temperature program were sharper, and the TMA peak was better separated. Therefore, it was chosen to continue with the standard solvent GC temperature program, see Appendix 8.2 for all the settings of this program. In addition, the program was 21.6 instead of 29 minutes, which is less time consuming. The chromatogram with the standard temperature program is shown in figure 6.
Figure 6: GC-MS chromatogram of commercially available alkyd resin, where the monomers phthalic anhydride (peak 4), Polyol pentaerythritol (peak 7), trimellitic anhydride (peak 9), and fatty acids are identified.
2.4 Conclusion
For a commercially available alkyd resin, it was possible to identify the monomers by NMR and GC-MS after a microwave digestion with TMAH in 25 wt% methanol at 120 °C with a ramp and hold time of 5 minutes. These methods could now be used and optimized to identify the monomers of polyamides and polyamide-modified alkyd resin.
3. Polyamide identification and quantification
At first test methods from literature and AkzoNobel that might work were tested. From this investigation the most promising method was chosen and further explored.
3.1 Test method exploration
First, the method of an earlier student here at AkzoNobel using micro-reaction GC-MS was future explored. A LC column was emptied and with this a kind of microreactor was built at larger scale to create a large volume reaction vessel. A LC column was chosen as these can easily handle pressure up to 400 bar. The commercially available polyamide samples were dissolved in TMAH 25wt% MeOH. The oven temperature, time, and amount of TMAH (25% in methanol) were varied. In each chromatogram many different small amine peaks were observed. However, the large amount of substituents and variations, did not help to identify the amine used in polyamide. Often, peaks from 2-methoxy-N,N-dimethylamine, 2-dimethylaminoethanol, N,N,N’-trimethyl-diaminoethane, N,N,N’,N’-tetramethyl-1,2-diaminoethane, 1,4-dimethylpiperazine were obtained. The first peak, 2-methoxy-N,N-dimethylamine, can be 1,2-ethanediamine after an elimination reaction of NH3. 2-dimethylaminoethanol and N,N,N’-trimethyl-diaminoethane can be reaction products of 1,2-ethandiamine after substitution of OH or O-CH3. N,N,N’,N’-tetramethyl-diaminoethane is actually methylated 1,2-ethanediamine. The chromatograms are shown in Appendix 8.3 figure 25. After changing several parameters, it was not possible to obtain one amine peak after digestion, which was required for quantification. It is important to note that the metal of the column activates these reactions, so glassware would be more optimal. 20 Therefore, other methods have been looked into.
The method above was not tested using microwave digestion. Therefore, the method which was used for determining the alkyd resin was tested to identify the amine. It would have be nice if polyamide would also broke down with TMAH (25 wt% in methanol). However, even after changing several parameters, solvents, and hold times it was not possible to use this method for the identification of polyamides.
An article in the literature was found written by Zhang et al. who did a study on the analysis composition of semi-aromatic copolyimide via 13 C-NMR. 21 Here polyamides were hydrolyzed and analyzed by 13C-NMR. Hydrolysis took 24 hours in an oven at 140 °C. In this experiment, the microwave was used to speed up the hydrolysis into 15 minutes. Here, a digestion solvent of HBr and TFA was used. The method of Zhang et al.
monomers. When using higher temperatures in the microwave, the sample did not dissolve anymore for analysis by NMR or GC-MS. Nevertheless, infrared (IR) spectra showed that the sample contained amide. Consequently, amine has been detected in the GC-MS, but this was not convincing enough to further explore this method completely.
Following the complete article, an ampoule was filled with 0.15 g sample, 3 mL HBr, 1.5 mL TFA, and sealed with a flame. The ampoule was placed in the oven for 24 hours at 140 °C and then measured by NMR and GC-MS. The GC-MS chromatogram showed amine peaks around t = 1.2 min in front of the solvent peak. The peaks were minimal, the article did not fully describe the sample preparation and the results from the article did not match. Also, the analysis by NMR did not show any amine peak. Therefore, this method was not further explored. Another option was to raise temperature in the oven to 300 C. However, the glass ampoules broke and also NMR tubes could not handle the pressure.
It was still not possible to identify the amine peak of polyamide after microwave digestion by NMR or GC-MS because of non-soluble solids in the vessel. Then, another method found in literature by Colombini et al. was tested. 22 In this method, a reagent of 6M HCl in water was used during microwave digestion. The method was as follows; weight 0.2 g of commercially available polyamide 1 in a vessel, add 0.5 mL 6M HCl (in water) and place the vessel in the microwave. The temperature was set at 200 °C, ramp time of 5 min, hold time of 10 min, and the stirring speed was set on high. After the digestion method the samples were analyzed by GC-MS, however, no amine was obtained. Then several parameters were varied, such as concentration of the HCl, temperature, and holding times. Yet, HCl was dissolved in water, but also other solvents were tested such as THF, IPA, and pyridine. A summary of all the tested parameters are shown in table 7.
Table 7: An overview of all methods tested for the identification of the monomer composition of polyamides.
Digestion solvents
Temperature (C)
Hold times Concentration
(M) TMAH in 25 wt% MeOH 120, 140, 160, 180, 200, 220 10, 20, 30, 45, 60 TMAH in 25 wt% H2O 120, 140, 160, 180, 200, 220 10, 20, 30, 45,60 TMAH in 25
wt% DMSO 120, 140, 160,180, 200, 220 10, 20, 30, 45,60 time optimal10 min hold HBr and TFA 140, 160, 180, 200, 210 HCl in water 120, 140, 160, 180, 200 1M, 2M, 3M, 4M,5M, 6M HCl in DMSO 120, 140, 160 1M, 2M, 3M, 4M, 5M, 6M HCl in THF 120, 140, 160, 180 1M, 2M, 3M, 4M, 5M, 6M HCl in pyridine 120, 140, 160,180 1M, 2M, 3M, 4M,5M, 6M HCL in IPA 120, 140, 160, 1M, 2M, 3M, 4M, 4M optimal,
5M, 6M 3M or lower: very viscous
solution
Temperature (C)
Time (hours) Concentration
Column; TMAH in 25
wt% MeOH
250, 270, 300 1, 2, 5, 8 100 mg sample in 0.75, 1, 1.5 mL
Also found in literature was the use of hexafluorisopropanol in a volume ratio of 3:1 hexaflourisopropanol (HFIP) to chloroform-d (CDCL3) for the direct monomer composition with 13C-NMR. 23, 21 This method was not investigated due to the very expensive solvent.
3.2 Polyamide identification
In this chapter, the polyamide identification is done by the most promising test method of using a reagent of 4M HCl in IPA during a microwave digestion. The amines were analyzed by IR, GC-MS, DIMS and NMR.
3.2.1 Chemicals
The following chemicals were used for the sample preparation: a commercially available polyamide 1, obtained from Arizona Chemicals. Hydrochloric acid, fuming, 37% in water (CAS: 7647-01-0), potassium hydroxide (CAS: 1310-58-3), and deuterated water (CAS: 7789-20-0) were obtained from Acros Organics. Isopropanol (CAS:67-63-0), methanol (CAS: 67-56-1), deuterated methanol (CAS: 881-98-3), chloroform (CAS: 67-66-3), water (CAS: 7732-18-5) were obtained from VWR. Deuterated chloroform (CAS: 865-49-6) and deuterated dimethyl sulfoxide (CAS: 2206-27-1) were obtained from Sigma Aldrich. Tetrahydrofuran (CAS: 109-99-9) was obtained from Fisher Scientific.
3.2.2 GC-MS method
0.2 g of commercially available polyamide 1 was dissolved in 0.5 mL 4M HCL (in isopropanol) in a vessel and placed in the microwave Discover SP-DTM clinical (CEM). The digestion time consisted a ramp time of 5 minutes and hold time of 10 minutes at a temperature of 150 °C. The speed of stirring during the digestion was high. The top and bottom layers were separated and both an excess of 10% KOH was added. For the analysis an Agilent 7890A gas chromatograph was used. The sample was separated on the GC with a non-polar stationary phase column CP9151 from Agilent (dimensions 0.25 m x 30 m x 0.25 mm) and analyzed with the MS Agilent 5975C inert XL EI/CI MSD with triple axis detector. The inlet source was 280 °C and the inlet pressure was set on 55.8 kPa with a flowrate of 1 mL/min and a split ratio of 1:25. The carrier gas is helium, see Appendix 8.2 for all settings.
3.2.3 NMR method
0.2 g of commercially available polyamide 1 was dissolved in 0.5 mL 4M HCl (in isopropanol) in a vessel and placed in the microwave Discover SP-DTM clinical (CEM). The digestion time consisted a ramp time of 5 minutes and a hold time of 10 minutes at a temperature of 150 °C. The speed of stirring during the digestion was high. The top and bottom layers were separated, and both dissolved in 0.6 mL DMSO-d6 and transferred in an NMR-tube. For the analysis a Bruker Avance III NMR spectrometer 400 MHz Ultrashield was used. A proton NMR (16 scans), 13C-NMR (2048 scans), 13 C-DEPT-135 (256 scans) spectra were included.
3.2.4 Identification by GC-MS
The sample after the microwave digestion was never homogeneous and the two layers needed to be separated before analysis. Multiple experiments were done by testing different solvents to make the sample homogeneous. Commercially available polyamide 3 was used for this experiment and deuterated solvents were tested. In earlier experiments pyridine, THF, and DMSO were already tested to dissolve the sample without success. The used solvents were methanol-d4, tertrahydrofuran-d8, 1,1-2,2-tetrachloroethane-d2 (TCE), 2-propanol-tertrahydrofuran-d8, ethanol-d6, chloroform-d1, and dimethyl sulfoxide-d6. Finally, 0.3 mL (0.6 mL for TCE and Chloroform) was added to the samples and placed in an ultrasonic bath for 30 minutes. After half an hour, no homogeneous solution was obtained. To conclude, no solvent could obtain a homogeneous solution after digestion. From now on, the two layers were always separated before analysis by GC-MS and NMR. It is important to note that isopropanol was divided over two layers.
The method of Colombini et al. was tested using different parameters, such as concentration of the HCl, temperature, and holding times. Also, HCl was dissolved in water, THF, IPA, and pyridine. Then, after the digestion of 4M HCl in IPA, two liquid layers were observed, and both were measured with IR. In the yellow liquid top layer, amide peaks at 1556 and 1636 cm-1 and acid peaks at 1713 and 3365 cm-1 were observed, see figure 7. Also, an amine group was observed in the IR database of AkzoNobel. In figure 8, the IR spectrum of the bottom colorless liquid layer is shown. Here, also amide peaks at 1556 and 1636 cm-1 were observed. The amide peak at 1636 cm-1 had a stronger absorbance in the bottom layer due to amine salt formation. Amine salt also give a peak around 1640 cm-1, this means that HCl was obtained in the bottom layer. The peak at 1640 cm-1 in relation to the peak at 1556 cm-1 is higher in the bottom layer than in the top layer, see blue lines in figure 7 and 8.
Figure 7: IR spectrum of the top layer of commercially available polyamide 1 after a digestion with 4M HCl in IPA (black). The red spectrum represents the commercially available polyamide 1 before digestion. The amide peaks are observed at 1556, 1636 and 3365 cm-1.
Figure 8: IR spectrum of the bottom layer of commercially available polyamide 1 after a digestion with 4M HCl IPA (black). The red spectrum represents the commercially available polyamide 1 before digestion. The amide peaks are observed at 1556, 1636, and 3352 cm-1.
To conclude, in the top layer, fatty acids were detected and in the bottom layer amine was observed.
Top layer = Fatty acids + IPA
Bottom layer = Amine + HCl + IPA
Finally, the two separated layers of commercially available polyamide 1 were analyzed by GC-MS. THF stabilized with butylhydroxytoluene was added to the bottom layer to dilute the sample and gain more volume. The THF peak in the chromatogram can be found around 1.6 min and butylhydroxytoluene (BHT) at 9 minutes. In the bottom layer, 1,2-ethanediamine was observed around 1.9 min and ethylamine at 11 min according to the database, but this peak is most probably protonated 1,2-ethanediamine (figure 9). In the top layer, no amine peaks were observed only fatty acids were obtained, and the chromatogram is shown in appendix 8.4 figure 26.
Figure 9: The bottom layer of commercially available polyamide 1 after digestion with 4M HCl in IPA in the microwave at a temperature of 150 °C and a hold time of 10 minutes. At t= 1.9 min 1,2-ethanediamine is observed and at 11 min ethylamine.
In figure 9 a big bump around the 11 minutes of protonated 1,2-ethanediamine was obtained, which was part of 1,2-1,2-ethanediamine. This was probably due to the low pH of HCl, which was also obtained in the chromatogram at 12.5 minutes. At pH 1-2, the amine is protonated (NH4+), while at pH 12 the amine is a base and has no charge (NH3). By adding KOH before analysis with the GC-MS, the amine will be deprotonated. The peak will probably move to the 1,2ethanediame peak at 1.8 minutes because when amine is deprotonated, it has no interaction with the silanol group of the column anymore. KOH was a strong base and neutralized the HCl. It was chosen to use KOH instead of NaOH, because KOH was more non-polar and will dissolved more easily in the alcohols. In figure 10 the chromatogram is shown after neutralizing the solution before analysis by GC-MS.
Figure 10: Chromatogram of the bottom layer of commercially available polyamide 1 after digestion with 4M HCl in IPA in the microwave at a temperature of 150 °C and a hold time of 10 minutes. Then 10% excess KOH was added to neutralize the HCl. 1,2-ethanediamine was obtained at 1,8 minutes and no bump around the 11 min was observed.
To conclude, monomers of commercially available polyamide 1 were identified using a microwave digestion with 4M HCl in IPA at 150 °C with a ramp time of 5 min and hold time of 10 minutes. The stirring speed was set on high. The sample was then digested into two layers. The two layers needed to be separated and dissolved in 10% excess of KOH to neutralize the solution from the low pH of HCl. For the identification of the monomers, the bottom layer of the samples was analyzed by GC-MS. At 1.8 minutes a 1,2-ethanediamine peak was detected.
Other polyamides were prepared and analyzed by GC-MS using the optimized method. Four polyamides were obtained from the polymer laboratory at AkzoNobel Sassenheim. The polyamides are referred as CRG1, CRG2, CRG3 and CRG4. Also, two other unknown polyamides were obtained from Arizona Chemicals. The unknown polyamides are referred as commercially available polyamide 2 and 3. Furthermore, Nylon 66 was also measured. Subsequently, the method was tested to work for other polyamides. The samples were all prepared using the optimized method and measured by GC-MS. In table 8 the results of the analysis by GC-MS are shown. The amine peaks, 1,2-ethanediamine and/or 1,6-hexanediamine, were all detected in the bottom layer of the sample after digestion. One example of a GC-MS chromatogram of commercially available polyamide 2 is given in figure 11. Here, 1,2-ethanediamine was identified at 1.8 minutes and 1,6-hexanediamine at 5.8 minutes. It can be concluded that this method can identify polyamides in different samples.
Table 8: Results of the polyamide samples by GC-MS, whether the amine peaks is identified or not, using the optimized method.
Samples Amine peak identified Comments
CRG0001 top layer 1,6-hexanediamine (t=5.8 min) Very small peak, maybe not well separated from bottom
layer. CRG0001 bottom
layer 1,6-hexanediamine (t=5.8 min)
CRG0002 top layer
-CRG0002 bottom
layer 1,6-hexanediamine (t=5.8 min) No THF added, does notdissolve
CRG0003 top layer
-CRG0003 bottom
layer 1,6-hexanediamine (t=5.8 min) No THF added, does notdissolve
CRG0004 top layer
-CRG0004 bottom
layer 1,6-hexanediamine (t= 5.8 min) Commercially available polyamide 2 top layer -Commercially available polyamide 2 bottom layer 1,2-ethanediamine and 1,6-hexanediamine (t=1.8 min and
t=5.8 min) Commercially available polyamide 3 top layer -Commercially available polyamide 3 bottom layer 1,6-hexanediamine (t=5.8min)
Nylon 66 top layer - Temperature microwave:
180 °C Nylon 66 bottom
layer 1,6-hexanediamine (t=5.8 min) Temperature microwave:180 °C
Figure 11: Chromatogram of the bottom layer of commercially available polyamide 2 after digestion with 3M HCl in IPA in the microwave at a temperature of 160 °C and a hold time of 10 minutes. Then 10% excess KOH was added to neutralize the
acid. All these fatty acids were found in the chromatogram. In all probability, the monomers of the commercially available polyamides were 1,6-hexandiamine and/or 1,2-ethanediamine and sunflower oil. Dimeric acid was not detectable by GC-MS. The results are shown in table 9.
Table 9: Results of the monomers found in the commercially available polyamides after microwave digestion with 3M HCl in IPA and analysis by GC-MS.
Commercially available polyamide 1 Commercially available polyamide 2 Commercially available polyamide 3
1,2-Ethanediamine 1,2-Ethanediamine 1,6-Hexanediamine Octanoic acid* 1,6-Hexanediamine Octanoic acid* Decanoic acid* Myristic acid Decanoic acid*
Myristate* Palmitate* Myristate*
Palmitic acid Palmitic acid Palmitic acid Linoleic acid Linoleic acid Linoleic acid
Stearic acid Stearic acid Stearic acid
Oleic acid Oleic acid Oleic acid
Arachidic acid
3.2.5 Identification by NMR
The samples were also analyzed by NMR. The top and bottom layer were dissolved in 0.3 mL DMSO-d6. The amine peak was observed around 38 ppm in the bottom layer of commercially available polyamide 1. However, also a small amine peak was found in the top layer. All the fatty acids were detected in the top layer. In figure 12 a zoomed in 13C-NMR spectrum between 0-70 ppm is shown. For the total 13C-NMR spectrum see appendix 8.5 figure 27.
To confirm if the peak observed at 38 ppm was 1,2-ethaneamine, 2D NMR, such as HMBC-, HSQC-, and TOCSY-NMR measurements were done. In the HMBC-NMR spectrum, which shows the coupling between carbon and protons, showed that the peak at 3.1 ppm in the proton NMR correlated to the peak at 37 ppm in the 13C-NMR spectrum. Also, in TOCSY-NMR a correlation between the 1,2-ethanediamine peak (3.1 ppm), the water peak (4 ppm) and NH2 groups (8 ppm) were observed. In general, hydroxy and amine protons were very exchangeable with neighboring protons groups at room temperature. The 2D NMR spectra confirmed that
1,2-ethanediamine was observed at 37 ppm in 13C-NMR and at 3.1 ppm in
proton NMR. In figure 13, TOCSY-NMR spectrum is shown and the HMBC-and HSQC-NMR spectra are shown in appendix 8.5 figure 28 HMBC-and figure 29.
However, for the quantification an extraction step needs to be added to separate both layers.
Figure 13: A TOCSY-NMR spectrum of commercially available polyamide 1 is shown after the digestion with 4M HCl in IPA in the microwave with a temperature of 150 C and a hold time of 10 minutes. At 3.1 ppm the 1,2-ethanediamine is observed, water is around 4 ppm and amine group protons are at 8 ppm shown. Hydroxy and amine protons are very readily exchangeable with neighboring protons which confirms that the peak observed in the 13C-NMR (figure 12) is 1,2-ethanediamine.
Other polyamide samples were also prepared for analysis by NMR. However, the peak of 1,6-hexanediamine was observed under the peak of DMSO-d6 using DEPT-NMR. Therefore, it was not possible to see the 1,6-hexanediamine peak by NMR.
It was now possible to identify the polyamides by GC-MS and 1,2-ethanediamine by NMR after sample preparation. However, the sample after the microwave digestion was never homogeneous and needed to be separated before analysis. Therefore, an extraction step was added to the sample preparation. Zhang et al. used an extraction method using chloroform. 21 The sample was transferred to a beaker and adjusted to pH 12 with 40% NaOH (aq). The solution was extracted with chloroform (3x 30 mL) in a separation funnel. The organic layer was dried over anhydrous Na2SO4, filtered and evaporated. The sample was then analyzed with proton and 13C-NMR. The amine peak in H-NMR was expected to be around 3 ppm and 40 ppm in 13C-NMR. Then, HMBC-, HSQC- and COSY-NMR spectra were taken to confirm amine in the sample. After analyzing the sample with 2D NMR measurements, it was confirmed that the peak at 2.70 ppm in H-NMR and the peak at 41.91 ppm in 13C-NMR was non protonated amine. The HMBC-NMR spectra is shown in figure 14. In this spectra H-NMR is shown on the x-axis and 13C-NMR on the y-axis. The peak at 2.70 ppm in H-NMR couples to the peaks at 41.91, 32.90, and 26.60 ppm in the 13C-NMR spectrum. The two peaks at 41.91 ppm are a one band coupling and should also be in the middle at 2.70 ppm. These numbers also correspond to the theoretical chemical shifts, see table 10. Therefore, the 1,6-ethanediamine peak was confirmed by using 2D NMR. Other NMR spectra is shown in Appendix 8.5 figure 30-33. Using this extraction method, amine can be identified by NMR.
Figure 14: An HMBC-NMR spectrum of CRG4 is shown after digestion with 4M HCl in IPA in the microwave at a temperature of 150 °C and a hold time of 10 minutes. Then, the sample was extracted with a method of Zhang et al. using chloroform. The organic layer was measured and shown in this 2D spectrum. The 1,6-hexanediamine peaks was observed at 2.70 ppm on the x-axis and couples to the peaks at 41.91, 32.90, and 26.60 ppm in 13C-NMR on the y-axis.
Table 10: Structure formula and theoretical chemical shifts of 1,6-hexanediamine in H-, and 13 C-NMR.
1,6-hexanediamine
Number in structure formula Chemical shift in 13
C-NMR (ppm) Chemical shift inH-NMR (ppm)
2 and 7 42.27 2.91
3 and 6 32.04 1.36
4 and 5 27.11 1.22
Another extraction method was tested. After microwave digestion, 2 mL of deuterated water and 2 mL of deuterated chloroform were added, and the sample was shaken. The water layer was a white viscous (milky, soapy) solution, which was most probably an emulsion. It was not possible to measure the emulsion by NMR; the sample has to be homogeneous. Therefore, 1 mL was diluted in 2 mL deuterated methanol before analyzing by NMR. The amine would be expected in the water layer and fatty acids in the chloroform layer. The sample was analyzed with proton and 13C-NMR. In 13C-NMR a very small peak around 40 ppm was detected, but it was almost in the signal to noise ratio. A reason for that can be too low concentrations because of the extraction step, which also diluted the sample. The sample was also analyzed with TOCSY- and HSQC-NMR to confirm amine. These 2D NMR spectra confirmed that amine in the H-NMR spectrum was at 2.97 ppm and in 13C-NMR spectrum at 39.34 ppm. The H-, 13C, HSQC-, and TOCSY-NMR spectra are shown in Appendix 8.5 figure 34-37. It can be concluded that proton NMR can identify amine after the microwave digestion and an extraction step of water-d2/methanol-d4: chloroform-d1.
3.2.6 Identification of dimeric fatty acids
A typical dimeric fatty acid was used as one of the monomers in CRG samples. Dimeric fatty acid consists of linear, cyclic, and aromatic dimeric fatty acids. The structures are shown in figure 15. The typical isomer composition is 7% linear, 65% cyclic, and 18% aromatic. The remaining 10% is probably monomer or trimer fatty acid.
Figure 15: Structures of typical isomers of dimeric fatty acids. The typical composition is 7% linear (A), 65% cyclic (B, D), and 18% aromatic (C, E) dimer fatty acid. The other 10% is probably monomer or trimer fatty acid.
Identification of dimer fatty acids was not possible by GC-MS. Therefore, it was analyzed by Direct Infusion Mass Spectrometry (DIMS). The samples were prepared as described above. After microwave digestion, the top layer was diluted with MeOH 1:100 and injected. The commercially available dimeric fatty acid was used as a reference to determine the
optimal MS settings, see Appendix 8.6 table 16. The monoisotopic masses and common adduct ions of dimeric fatty acids in negative mode are shown in table 11.
Table 11: The monoisotopic mass (M) and m/z value of common adduct ions of dimeric fatty acids, [M-H]
and [M+Na-2H] -. Masses were calculated using Compass IsotopePattern.
Dimeric fatty acid M [M-H]- [M+Na-2H] -A (C38H74O4) 594.558 593.550 615.532 B (C38H72O4) 592.543 591,535 613.517 C (C38H70O4) 590.527 589,519 611.501 D (C38H66O4) 586.496 585,488 607.470 E (C38H64O4) 584.480 583.472 605.454
Then, the top layers of CRG2, CRG4, and commercially available polyamide 3 were measured by DIMS in negative mode. The masses [M-H]- of dimeric fatty acids were in all 3 samples observed in the spectra, see figure 16. This proves that dimeric fatty acids did not break during digestion and were still present in the samples. Dimeric fatty acids are non-polar and can cause viscosity problems. During these experiments multiple problems occur without a specific reason. The results were not always unambiguous, and it may have to do dimeric fatty acid.
3.3 Polyamide quantification
For the quantification of the polyamide samples, two calibration curves were made of reference samples 1,2-ethanediamine (5-35 mg/mL) and 1,6-hexanediamine (10-35 mg/mL). The samples were measured 3 times and a linear regression coefficient of >0.991 was obtained, see figure 17.
0 5 10 15 20 25 30 35 40 0.00E+00 2.00E+06 4.00E+06 6.00E+06 8.00E+06 1.00E+07 1.20E+07
Calibration curves
1,6-hexanediamine Time (min) A b u n d an ceFigure 17: Calibration curves of reference samples of diamine. The orange line refers to 1,2-ethanediamine with a concentration range of 5-35 mg/mL, y = 258573x + 320320, n = 3, r2
= 0.9917. The blue line refers to 1,6-hexanediamine with a concentration range of 10-35 mg/mL, y = 602998x – 381973, n = 3, r2 = 0.9917.
Then, the samples obtained from Polymer laboratory at AkzoNobel and the unknown polyamides obtained from Arizona Chemicals were tested. The samples were prepared as described in section 3.2.2 GC-MS method. The areas of the peaks were integrated and with the calibration curves, the concentration was calculated. The amine concentration of the samples from polymer laboratory was known, so a recovery was calculated. Very low recovery of 11-39% was obtained.
Table 12: The calculated concentration of amine in the polyamide samples is shown using the calibration curve (figure 17). Also, the recovery was calculated for the polyamides obtained from Polymer laboratory of AkzoNobel.
Sample Weight
(mg/mL) amine recipeTheoretical (%) Weight in reaction (mg/mL) Area Measured conc. (mg/mL) Recovery (%) CRG1 389.87 11.25 43.86 696896 4.83 11.01 CRG2 387.6 8.76 33.94 299830 4.20 12.37 CRG3 395.93 8.48 33.57 442365 4.43 13.18 CRG4 398.68 16.85 67.18 1422115 4 26.29 39.14 Found
available polyamide 2 ethanediamine Commercially available polyamide 3 393.02 1,6-hexanediamine 4666575 11.13 2.83
The samples obtained from Polymer laboratory at AkzoNobel were also analyzed by NMR. The samples were prepared as described in section 3.2.3 NMR method. After digestion, an extraction step of d:4 chloroform-d1 was performed. In the water-d2/methanol-d4 layer amine was observed. The ERETIC method was applied for quantification of the samples. Therefore, a reference sample of 1,6-hexanediamine was prepared in water-d2/methanol-d4 and measured. This was used as an internal calibrant. In H-NMR a peak of 1,6-hexanediamine was obtained around 3.35 ppm. For the reference sample, the weight in was 21.03 mg and molar mass of 116.21 g/mol. The peak area was selected, and a concentration of 180.97 mmol/L was calculated. Then, the CRG samples were measured. The amine peak was shifted to 2.93 ppm, and peak areas were selected. The number of 4 equivalent atoms was used. This resulted in a calculated concentration of amine per sample. The samples were 7.5 times diluted, with the assumption that IPA was divided equally in both layers. In table 13, the weight in of the CRG samples is given with the theoretical calculated amount of amine. The calculated concentration of amine per sample and recovery is given in table 14. A very low recovery of 1.96-3.16 % was obtained. Unfortunately, it was not possible to optimize the quantification of polyamides by NMR due to instrument issues.
Table 13: Theoretical amount of amine in the samples CRG1-4 were calculated.
Sample
s Weight in(mg) Percentage aminein sample (%) Amount amine (mg) n (mmol)
CRG1 411.14 11.25 46.25 0.398
CRG2 393.14 8.76 34.42 0.296
CRG3 428.74 8.48 36.35 0.313
CRG4 437.93 16.85 73.79 0.635
Table 14: Calculated concentration and recovery of samples CRG1-4.
Sample
s Concentration(mmol/L) n (mmol) Amount amine(mg) Recovery(%)
CRG1 1.039 0.00779 0.91 1.96
CRG2 1.249 0.00937 1.09 3.16
CRG3 1.017 0.00763 0.89 2.44
CRG4 1.714 0.0129 1.49 2.02
The recoveries obtained from GC-MS and NMR were extremely low. A possibility could be that the reaction was not fully completed, and therefore just a part of amine was obtained after digestion. To confirm this
theory, titrations were done to observe the conversion. The top layer was dissolved in 30 mL ethanol/xylene solution 1:1 and titrated with 0.1 M KOH/MeOH (VWR CAS:1310-58-3). For the bottom layer, KOH was added to obtain a pH of 7 and then dissolved in 30 mL acetic acid (>99.7% VWR CAS: 64-19-7). The titration was done by 0.1 M perchloric acid (HClO4) in anhydrous acetic acid (VWR CAS: 7601-90-3). Unfortunately, no titration curve was obtained of the bottom layer. The reason is unknown.
The top layer containing fatty acids was titrated with 0.1 M KOH/MeOH titer. The exact concentration was 0.1038 M. The titration curve had two equivalents, first the strong reagent of 4M HCl in IPA was compensated. The second equivalent was used to calculate the amount of fatty acids. First a reference sample of sunflower oil fatty acids was prepared and dissolved in 4M HCl in IPA. Theoretical 4M KOH, which is 38.53 mL of 0.1038 M KOH solution, needs to be titrated to neutralize 4M HCl and to get to the first equivalent point. Then, 58.06 mg was weight in, which was similar to 0.213 mmol sunflower oil fatty acid. The theoretically second equivalent using a 0.1038 M KOH solution, should be after 1.99 mL. After analysis, the first equivalent was obtained after 34.02 mL, see figure 18. Sunflower oil fatty acid is a neutral product, the composition of sunflower oil is not extract. Therefore, it may differ from the theoretical value. Then, CRG samples were measured and the conversion of fatty acids were calculated. A conversion of 8-33% was observed. However, also a few fatty acids were obtained in GC-MS chromatograms in the bottom layer, so this was just to get an indication of conversion. The calculated results are shown in table 15 and titration curves can be found in appendix 8.7 figure 38. It was noticed that the first equivalent of CRG samples were measured around 1-3 mL. This means that 95% of 4M HCl in IPA was reacted during digestion in the microwave. Therefore, it is possible that the samples just partly digested, because of the HCl was running low during digestion.
Figure 18: Titration curve of sunflower oil fatty acid. As a titer 0.1038 M KOH/MeOH was used. Equivalent point one was at 34.02 mL to neutralize the HCl and equivalent point was at 36.6 mL. Between eq1 and eq2 the converted amount of acid was measured. This means that 127.59% of acid was converted, see calculations in table 15.
Table 15: Calculated conversion of fatty acids in CRG samples during digestion with 4M HCL in IPA.
Weight in (mg)
M (g/mol)
Eq1 Eq2 Consumption
(mL) n acid (mmol COOH/g hars) n KOH (mmol KOH/g hars) Converte d (%) Ref. sunflower fatty acid 58.06 274.50 34.02 6 36.626 2.600 3.643 4.648 127.59 Weight in (mg) Mass (%) M (g/mol)
Eq1 Eq2 Consumption
(mL) n acid (mmol COOH/g hars) n KOH (mmol KOH/g hars) Converte d (%) CRG1 406.43 1.913 4.411 2.498 0.259 8.33 Dimeric fatty acid 88.75 570 3.114 Weight
in (mg) Mass(%) (g/mol)M Eq1 Eq2 Consumption(mL) COOH/g hars)n acid (mmol n KOH (mmolKOH/g hars) Converted (%)
CRG2 396.85 1.195 4.477 3.282 0.859 22.45 Dimeric fatty acid 69.1 570 2.425 Isononanic acid 22.14 158.24 1.399 Total: 3.824 Weight in (mg) Mass (%) M (g/mol)
Eq1 Eq2 Consumption
(mL) n acid (mmol COOH/g hars) n KOH (mmol KOH/g hars) Converte d (%) CRG3 402.06 1.361 4.391 3.030 0.782 18.89 Dimeric fatty acid 66.92 570 2.348 Sunflower oil fatty acid 24.6 274.50 1.792 Total: 4.140 Weight
3.4 Summary polyamide identification and quantification
The most promising method was tested using a reagent of 4M HCl in IPA during digestion. Samples were all identified by GC-MS and NMR. However, unambiguously results were not always obtained. Moreover, each amine reacted differently and needed some optimization. For example, nylon 6,6 was also tested with this method and 1,6-hexanediamine was found when the digestion method was done at 180 C. It was sometimes hard to understand how amine reacted. In addition, dimeric fatty acids can change the normality and expected results due to their big structures. Also, dimeric fatty acids were nonpolar and can cause viscosity problems. The easiest identification was by GC-MS, after preparing the sample, KOH was added, and the sample was measured. It was then simple to identify amine in the chromatogram. To determine the amine after analysis by NMR was a bit more challenging due to low sample concentrations, especially for 13C-NMR. However, proton NMR and 2D-NMR could confirm amine in the samples.
Quantification of polyamides samples was difficult. Very low recoveries were obtained by GC-MS and NMR using different extraction methods. A possibility was that the total conversion of amine was not obtained after digestion. Therefore, the samples were measured by titration to get more insight of total conversion of the reaction. For the bottom layer, no titration curve was obtained. The fatty acids in the top layer could be detected, a titration curve of 2 equivalent points was obtained. The total conversion was calculated for all samples and were between 8-33%. This was just an indication, because in GC-MS chromatograms of the bottom layer, some fatty acids were also observed. However, it does give more insight and understanding of the process. It also makes more sense that during quantification of polyamides such low recoveries were obtained. It was remarkable to see that 95% of reagent HCL in IPA was used up during the digestion step. It can be that the samples are therefore partly digested, because the HCl is running low. Therefore, quantification of polyamides needs to be further explored to obtain higher recoveries.
