NMR and GPC Analysis of Alkyd Resins: Influence of Synthesis Method, Vegetable Oil and Polyol Content

Alkyd resins are oil-based polymers that have been widely used for generations in the surface coating industry and beyond. Characterization of these resins is of high importance to understand the influence of its components on its behavior, compatibility with other resins, and final quality to ensure high durability. Here, NMR spectroscopy and GPC were used for characterizing differences in the chemical structure, molecular distribution, and dispersity between oil-based and fatty acid-based alkyd polymers made from sacha inchi and linseed oils. Sancha inchi (Plukentia volubilis L.) is a fruit-bearing plant native to South America and the Caribbean, and has a rich unsaturated fatty acid content. The effect of vegetable oil and polyol selection on the synthesis of alkyd resins for coating applications was analyzed. The influence of two different synthesis methods, monoglyceride and fatty acid processes, was also compared. Important structural differences were observed using NMR: one-dimensional spectra revealed the degree of unsaturated fatty acid chains along the polyester backbone, whereas, 2D NMR experiments facilitated chemical shift assignments of all signals. GPC analysis suggested that alkyd resins with homogeneous and high molecular weights can be obtained with the fatty acid process, and that resins containing pentaerythritol may have uniform chain lengths.


Introduction
Alkyd resins are polyesters modified using vegetable oils [1]. Conjugated carbon double-bonds from fatty acid chains, preferably catalyzed by metallic driers, afford the polymer the ability to generate radical autoxidation reactions [2], which facilitate its crosslinking and the formation of a dry film. This inherent characteristic gives alkyd resins the advantage of being single component coatings and, due to their oily content, they can better tolerate the presence of rust on the surface. In addition to being easy to apply, these vegetable oil-based coatings are cheap and have been successfully used in a variety of applications, including industrial, decorative, architectural or artistic coatings [3][4][5][6]. In recent years, they are being explored for self-healing applications to ensure the long-term durability of protective coatings [7,8]. Alkyd resins also have a great compatibility with other resins, and have been blended with other polymers to improve their performance [4].
The wide range of reagents that can be used for the synthesis of alkyd resins increase their versatility, and their structures and film properties can be tuned by their components [9]. Vegetable oils as fatty acid sources, polycarboxylic acids, such as phthalic anhydride (PA), and polyols, such as glycerol (GC) or pentaerythritol (PE), are commonly used for alkyd resin production [10]. Except for phthalic anhydride, alkyd polymers are

NMR Spectroscopy
NMR analysis was performed in the Department of Physical Chemistry and Microreaction Technology at the Technische Universität Ilmenau (Ilmenau, Germany) with a Bruker Avance 300 spectrometer. Samples were dissolved with deuterated chloroform (10 wt.%). Data analysis was performed using the MestReNova v12.0.4-22023 software.

Gel Permeation Chromatography
Gel Permeation Chromatography (GPC) was performed in the Department of Chemical Engineering at The Pennsylvania State University (USA). Molecular weights and molecular weight distribution (Mw/Mn) were determined with the 1260 Agilent Technologies gel permeation chromatograph (GPC) equipped with a refractive index detector (RID), a viscometer (VS), a light scattering detector (LS) and a multiwavelength detector (MWD), using an Agilent PLgel-MIXED-LC column. Chlorobenzene was used as the mobile phase at 0.5 mL/min and the sample concentration was~30 mg/mL. Universal GPC calibration curve was prepared with low polydispersed linear polystyrene standards (Agilent EasiVial PS-M, Agilent, Santa Clara, CA, USA).

1D 1 H-NMR Spectra
Similar 1 H NMR (300 MHz, chloroform-d) spectra were obtained for all alkyd resins. Figure 1 shows the 1 H NMR spectra that contains thirteen signals typical of an alkyd resin Polymers 2023, 15, 1993 4 of 14 classified with letters (A) to (M), starting from the high field region. Possible location of protons corresponding to each peak in alkyd resin structures (representing possible repeating units) is also identified in Figure 1. The unmarked peak at δ 7.28 ppm corresponds to chloroform-d. The assignment of 1 H-NMR chemical shifts in FAS3, AS3, AL3, AS1 and AL1 alkyd resins are summarized in Table 2. For the NMR resonance description, the "NMR Guidelines for ACS Journals" standard was used. For example, AS3 can be summarized as follows: GPC calibration curve was prepared with low polydispersed linear polystyrene standards (Agilent EasiVial PS-M, Agilent, Santa Clara, CA, USA).

1D 1 H-NMR Spectra
Similar 1 H NMR (300 MHz, chloroform-d) spectra were obtained for all alkyd resins. Figure 1 shows the 1 H NMR spectra that contains thirteen signals typical of an alkyd resin classified with letters (A) to (M), starting from the high field region. Possible location of protons corresponding to each peak in alkyd resin structures (representing possible repeating units) is also identified in Figure 1. The unmarked peak at δ 7.28 ppm corresponds to chloroform-d. The assignment of 1 H-NMR chemical shifts in FAS3, AS3, AL3, AS1 and AL1 alkyd resins are summarized in Table 2. For the NMR resonance description, the "NMR Guidelines for ACS Journals" standard was used. For example, AS3 can be summarized as follows:   The low field region of the spectrum at δ = 6.5-8.0 ppm aromatic ring protons, peaks (L) and (M), from the phthalic ester moieties. The resonance at about δ = 5.4 ppm, peak (K), was assigned to the vinylic protons. Intermediate signals δ = 3.4-4.8 ppm, composed of peaks (H), (I) and (J), correspond to the protons assigned to the polyol(s) CH 2 groups. The resonance at about δ = 2.8 ppm, peak (G), was assigned to aliphatic CH 2 groups shielded by two neighboring vinyl groups. Remaining peaks in the aliphatic region δ = 0-2.6 ppm correspond to sp 3 C bound protons assigned to the fatty acid chains, as shown in Table 2. The proton resonance assignment of all peaks is summarized in Table 2 and they align with data in the literature [29,32,33,39]. Previous studies of alkyd resins were reported by Boruah et al. [21], Spyros [29], Chiplunkar and Pratap [31], Rämänen and Maunu [33], and Glenn et al. [34].
An expansion of the 1 H spectra polyol region of the alkyd resins is shown in Figure 2. Through a visual comparison of all the spectra, a clear difference can be seen between alkyds prepared with glycerol (e.g., AS1, AL1) and those prepared with pentaerythritol (e.g., AS3, AL3, FAS3). As specified in Table 2, signal (H) correspond to methylene protons of the CH 2 OH groups. The degree of branching depends on the esterification degree of the polyol unit [33]. However, PE has a higher degree of OH-functionality than GC. Because of the steric differences, there is a higher probability that more unreacted CH 2 -OH groups remain in the PE structure in comparison to GC. This assumption is supported by the observation that the intensity of peak (H) is more intense in case of the PE alkyds (AL3, AS3, FAS3) compared to the GC alkyds (AL1, AS1). Peak (I), assigned to the esterified CH 2 -O-moieties of the PE-containing alkyd resins, has higher signal intensity because of the greater amounts of PE in the batch. On the other hand, peak (J)'s intensity is comparable for all samples, because the same amount of phthalic anhydride was used for the synthesis of all alkyd resins variants.  Figure 1) on alkyd resins. The resonances were measured in deuterated chloroform solution and assigned to the polymer structure units [21,29,31,33,34].  The low field region of the spectrum at δ = 6.5-8.0 ppm aromatic ring protons, peaks (L) and (M), from the phthalic ester moieties. The resonance at about δ = 5.4 ppm, peak (K), was assigned to the vinylic protons. Intermediate signals δ = 3.4-4.8 ppm, composed of peaks (H), (I) and (J), correspond to the protons assigned to the polyol(s) CH2 groups. The resonance at about δ = 2.8 ppm, peak (G), was assigned to aliphatic CH2 groups shielded by two neighboring vinyl groups. Remaining peaks in the aliphatic region δ = 0-2.6 ppm correspond to sp 3 C bound protons assigned to the fatty acid chains, as shown in Table 2. The proton resonance assignment of all peaks is summarized in Table 2   The low field region of the spectrum at δ = 6.5-8.0 ppm aromatic ring protons, peaks (L) and (M), from the phthalic ester moieties. The resonance at about δ = 5.4 ppm, peak (K), was assigned to the vinylic protons. Intermediate signals δ = 3.4-4.8 ppm, composed of peaks (H), (I) and (J), correspond to the protons assigned to the polyol(s) CH2 groups. The resonance at about δ = 2.8 ppm, peak (G), was assigned to aliphatic CH2 groups shielded by two neighboring vinyl groups. Remaining peaks in the aliphatic region δ = 0-2.6 ppm correspond to sp 3 C bound protons assigned to the fatty acid chains, as shown in Table 2. The proton resonance assignment of all peaks is summarized in Table 2   The low field region of the spectrum at δ = 6.5-8.0 ppm aromatic ring protons, peaks (L) and (M), from the phthalic ester moieties. The resonance at about δ = 5.4 ppm, peak (K), was assigned to the vinylic protons. Intermediate signals δ = 3.4-4.8 ppm, composed of peaks (H), (I) and (J), correspond to the protons assigned to the polyol(s) CH2 groups. The resonance at about δ = 2.8 ppm, peak (G), was assigned to aliphatic CH2 groups shielded by two neighboring vinyl groups. Remaining peaks in the aliphatic region δ = 0-2.6 ppm correspond to sp 3 C bound protons assigned to the fatty acid chains, as shown in Table 2. The proton resonance assignment of all peaks is summarized in Table 2   The low field region of the spectrum at δ = 6.5-8.0 ppm aromatic ring protons, peaks (L) and (M), from the phthalic ester moieties. The resonance at about δ = 5.4 ppm, peak (K), was assigned to the vinylic protons. Intermediate signals δ = 3.4-4.8 ppm, composed of peaks (H), (I) and (J), correspond to the protons assigned to the polyol(s) CH2 groups. The resonance at about δ = 2.8 ppm, peak (G), was assigned to aliphatic CH2 groups shielded by two neighboring vinyl groups. Remaining peaks in the aliphatic region δ = 0-2.6 ppm correspond to sp 3 C bound protons assigned to the fatty acid chains, as shown in Table 2. The proton resonance assignment of all peaks is summarized in Table 2    The low field region of the spectrum at δ = 6.5-8.0 ppm aromatic ring protons, peaks (L) and (M), from the phthalic ester moieties. The resonance at about δ = 5.4 ppm, peak (K), was assigned to the vinylic protons. Intermediate signals δ = 3.4-4.8 ppm, composed of peaks (H), (I) and (J), correspond to the protons assigned to the polyol(s) CH2 groups. The resonance at about δ = 2.8 ppm, peak (G), was assigned to aliphatic CH2 groups shielded by two neighboring vinyl groups. Remaining peaks in the aliphatic region δ = 0-2.6 ppm correspond to sp 3 C bound protons assigned to the fatty acid chains, as shown in Table 2. The proton resonance assignment of all peaks is summarized in Table 2    The low field region of the spectrum at δ = 6.5-8.0 ppm aromatic ring protons, peaks (L) and (M), from the phthalic ester moieties. The resonance at about δ = 5.4 ppm, peak (K), was assigned to the vinylic protons. Intermediate signals δ = 3.4-4.8 ppm, composed of peaks (H), (I) and (J), correspond to the protons assigned to the polyol(s) CH2 groups. The resonance at about δ = 2.8 ppm, peak (G), was assigned to aliphatic CH2 groups shielded by two neighboring vinyl groups. Remaining peaks in the aliphatic region δ = 0-2.6 ppm correspond to sp 3 C bound protons assigned to the fatty acid chains, as shown in Table 2. The proton resonance assignment of all peaks is summarized in Table 2    The low field region of the spectrum at δ = 6.5-8.0 ppm aromatic ring protons, peaks (L) and (M), from the phthalic ester moieties. The resonance at about δ = 5.4 ppm, peak (K), was assigned to the vinylic protons. Intermediate signals δ = 3.4-4.8 ppm, composed of peaks (H), (I) and (J), correspond to the protons assigned to the polyol(s) CH2 groups. The resonance at about δ = 2.8 ppm, peak (G), was assigned to aliphatic CH2 groups shielded by two neighboring vinyl groups. Remaining peaks in the aliphatic region δ = 0-2.6 ppm correspond to sp 3 C bound protons assigned to the fatty acid chains, as shown in Table 2. The proton resonance assignment of all peaks is summarized in Table 2    The low field region of the spectrum at δ = 6.5-8.0 ppm aromatic ring protons, peaks (L) and (M), from the phthalic ester moieties. The resonance at about δ = 5.4 ppm, peak (K), was assigned to the vinylic protons. Intermediate signals δ = 3.4-4.8 ppm, composed of peaks (H), (I) and (J), correspond to the protons assigned to the polyol(s) CH2 groups. The resonance at about δ = 2.8 ppm, peak (G), was assigned to aliphatic CH2 groups shielded by two neighboring vinyl groups. Remaining peaks in the aliphatic region δ = 0-2.6 ppm correspond to sp 3 C bound protons assigned to the fatty acid chains, as shown in Table 2. The proton resonance assignment of all peaks is summarized in Table 2   The low field region of the spectrum at δ = 6. The resonance at about δ = 2.8 ppm, peak (G), was assigned to aliphatic CH2 groups shielded by two neighboring vinyl groups. Remaining peaks in the aliphatic region δ = 0-2.6 ppm correspond to sp 3 C bound protons assigned to the fatty acid chains, as shown in Table 2. The proton resonance assignment of all peaks is summarized in Table 2   The low field region of the spectrum at δ = 6. The resonance at about δ = 2.8 ppm, peak (G), was assigned to aliphatic CH2 groups shielded by two neighboring vinyl groups. Remaining peaks in the aliphatic region δ = 0-2.6 ppm correspond to sp 3 C bound protons assigned to the fatty acid chains, as shown in Table 2. The proton resonance assignment of all peaks is summarized in Table 2   The low field region of the spectrum at δ = 6.5-8.0 ppm aromatic ring protons, peaks (L) and (M), from the phthalic ester moieties. The resonance at about δ = 5.4 ppm, peak (K), was assigned to the vinylic protons. Intermediate signals δ = 3.4-4.8 ppm, composed of peaks (H), (I) and (J), correspond to the protons assigned to the polyol(s) CH2 groups. The resonance at about δ = 2.8 ppm, peak (G), was assigned to aliphatic CH2 groups shielded by two neighboring vinyl groups. Remaining peaks in the aliphatic region δ = 0-2.6 ppm correspond to sp 3 C bound protons assigned to the fatty acid chains, as shown in Table 2. The proton resonance assignment of all peaks is summarized in Table 2   The low field region of the spectrum at δ = 6.5-8.0 ppm aromatic ring protons, peaks (L) and (M), from the phthalic ester moieties. The resonance at about δ = 5.4 ppm, peak (K), was assigned to the vinylic protons. Intermediate signals δ = 3.4-4.8 ppm, composed of peaks (H), (I) and (J), correspond to the protons assigned to the polyol(s) CH2 groups. The resonance at about δ = 2.8 ppm, peak (G), was assigned to aliphatic CH2 groups shielded by two neighboring vinyl groups. Remaining peaks in the aliphatic region δ = 0-2.6 ppm correspond to sp 3 C bound protons assigned to the fatty acid chains, as shown in Table 2. The proton resonance assignment of all peaks is summarized in Table 2   The low field region of the spectrum at δ = 6.5-8.0 ppm aromatic ring protons, peaks (L) and (M), from the phthalic ester moieties. The resonance at about δ = 5.4 ppm, peak (K), was assigned to the vinylic protons. Intermediate signals δ = 3.4-4.8 ppm, composed of peaks (H), (I) and (J), correspond to the protons assigned to the polyol(s) CH2 groups. The resonance at about δ = 2.8 ppm, peak (G), was assigned to aliphatic CH2 groups shielded by two neighboring vinyl groups. Remaining peaks in the aliphatic region δ = 0-2.6 ppm correspond to sp 3 C bound protons assigned to the fatty acid chains, as shown in Table 2. The proton resonance assignment of all peaks is summarized in Table 2 and they align with data in the literature [29,32,33,39]. Previous studies of alkyd resins were reported by Boruah et al. [21], Spyros [29], Chiplunkar and Pratap [31], Rämänen and Maunu [33], and Glenn et al. [34].

Peak
Polymers 2023, 15, 1993 6 of 14 by the observation that the intensity of peak (H) is more intense in case of the PE alkyd (AL3, AS3, FAS3) compared to the GC alkyds (AL1, AS1). Peak (I), assigned to the ester fied CH2-Omoieties of the PE-containing alkyd resins, has higher signal intensity becaus of the greater amounts of PE in the batch. On the other hand, peak (J)'s intensity is com parable for all samples, because the same amount of phthalic anhydride was used for th synthesis of all alkyd resins variants. The integrated peak area (M) at δ 7.71 ppm, corresponding to two protons attache to the aromatic ring of the phthalic anhydride structure, was used for integral normaliza tion of the full spectrum. The comparison of different alkyd resin samples revealed diffe ent relative proportions of polyol and fatty acid chain proton integrals. We assume tha that all polymer samples had the same amount of phthalic anhydride, because the sam relative amount of phthalic acid was used for the preparation of each batch. Peaks (H), ( and (J) from the polyol region were chosen to make comparisons between the differen samples. Peaks (A) and (B) were also selected, since they are representative signals of th linoleic and linolenic fatty acids, or the main components of vegetable oils used for alky synthesis. Other peaks were not taken into account because they might be present in similar proportion in every possible structure of a repeating unit of an alkyd polymer.
Relative normalized integrated areas of the characteristic signals corresponding t the polyol and fatty acid chain protons are illustrated in Figure 3. Relative proportions o the polyol protons corroborated the visual comparison made with the 1 H-NMR spectra Protons in the polyol region are in greater proportion in PE-based alkyd resins. It was als noted that PE-based resins contain more fatty acid chains, due to the highest functionalit of pentaerythritol. In the high field region, sacha inchi-based resins have more termina methyl groups found in a linoleic acid (omega-6) structure (peak (A) from Figure 1) tha linseed-based resins (Figure 3). It has been reported that sacha inchi has a similar omega 3 fatty acid composition to linseed oil; though, its omega-6 fatty acid content is higher [22 A higher degree of unsaturation, i.e., large quantity of double bonds, could enhance resi The integrated peak area (M) at δ 7.71 ppm, corresponding to two protons attached to the aromatic ring of the phthalic anhydride structure, was used for integral normalization of the full spectrum. The comparison of different alkyd resin samples revealed different relative proportions of polyol and fatty acid chain proton integrals. We assume that that all polymer samples had the same amount of phthalic anhydride, because the same relative amount of phthalic acid was used for the preparation of each batch. Peaks (H), (I) and (J) from the polyol region were chosen to make comparisons between the different samples. Peaks (A) and (B) were also selected, since they are representative signals of the linoleic and linolenic fatty acids, or the main components of vegetable oils used for alkyd synthesis. Other peaks were not taken into account because they might be present in a similar proportion in every possible structure of a repeating unit of an alkyd polymer.
Relative normalized integrated areas of the characteristic signals corresponding to the polyol and fatty acid chain protons are illustrated in Figure 3. Relative proportions of the polyol protons corroborated the visual comparison made with the 1 H-NMR spectra. Protons in the polyol region are in greater proportion in PE-based alkyd resins. It was also noted that PE-based resins contain more fatty acid chains, due to the highest functionality of pentaerythritol. In the high field region, sacha inchi-based resins have more terminal methyl groups found in a linoleic acid (omega-6) structure (peak (A) from Figure 1) than linseed-based resins (Figure 3). It has been reported that sacha inchi has a similar omega-3 fatty acid composition to linseed oil; though, its omega-6 fatty acid content is higher [22]. A higher degree of unsaturation, i.e., large quantity of double bonds, could enhance resin drying properties and, therefore, the hardness of cured resins. The development of extensive cross-linking would lead to the obtainment of a better quality product [29,40].
Polymers 2023, 13, x FOR PEER REVIEW 7 of 15 drying properties and, therefore, the hardness of cured resins. The development of extensive cross-linking would lead to the obtainment of a better quality product [29,40].

2D NMR Spectra
Sample AS3 was analyzed using different 2D NMR experiments in order to investigate the structure in greater detail. The 1 H-1 H COSY-90 (Figure 5a), 1 H-1 H TOCSY (Figure  5b), and 1 H-1 H ROESY (Figure 5c) correlation spectra were recorded. The analysis revealed clearly the coupling of peak (K) with peaks (E) and (G), located in the fatty acid chain

2D NMR Spectra
Sample AS3 was analyzed using different 2D NMR experiments in order to investigate the structure in greater detail. The 1 H-1 H COSY-90 (Figure 5a), 1 H-1 H TOCSY (Figure 5b), and 1 H-1 H ROESY (Figure 5c) correlation spectra were recorded. The analysis revealed clearly the coupling of peak (K) with peaks (E) and (G), located in the fatty acid chain (Figure 5a). Moreover, coupling of proton peaks (C), (D) and (F) were confirmed. It was corroborated that protons from terminal methyl groups of linoleic acid chain correspond to peak (A), as it has a strong coupling with protons from peak (C). Thus, peak (B), protons from terminal methyl groups of linolenic acid chain, showed a crosspeak with signal (E) in the COSY and ROESY measurements. It should be noted that methylene protons of peak (E) are also close to those of peak (C). The TOCSY-2D spectra (Figure 5b) additionally confirmed that protons of unsaturated carbons identified as peak (K), correlates with protons from internal methylene groups of the aliphatic chains (peak (C)).
The 2D 1 H-13 C HMQC spectra ( Figure 6) were also acquired for AS3 sample. Figure 6a shows the expanded spectra containing identified 1 H NMR peaks (A) to (G) (Figure 1), located in the δ 0.6 to 3.0 ppm region, and carbons C7 to C12 that appeared from δ 10 to 40 ppm in the 13 C NMR spectra (Figure 4). Proton peaks (A) and (B) showed the expected connectivity with carbon peaks around δ 13.8-14.3 ppm, C12 identified in Figure 4, which represent terminal methyl groups on fatty acid chains.
Heteronuclear coupling was used for the identification of the different 13 C NMR resonances to the methylene protons of the fatty acid chains. Here, peak (C) has a cross correlation with the carbon signals around δ = 29.1-29.7 ppm (identified in Figure 4 as carbon C9). The correlation of the proton peak at δ = 1.61ppm (peak (D)) with a carbon C8 peak at δ = 24.9 ppm, is attributed to the methylene groups located next to -CH 2 COOgroup. Proton peak (E) has a cross correlation with carbon C10 peak at δ = 34.1 ppm. Other cross correlations observed in this spectrum originated from coupling between the couples proton peak (F)-carbon C7 peak, and proton peak (G)-carbon C11 peak, that appeared at δ 34.2 ppm and δ 25.5-25.6 ppm, respectively.
The 1 H-13 C HMQC spectra of the polyol and aromatic regions of an alkyd resin are presented in Figure 6b. Detected signals of the expanded spectra were found in the range of δ 4.0-8.0 ppm in the 1 H NMR experiment, which included peaks (I) to (M) (Figure 1). It should be noted that peak (H) showed no clear correlation with a carbon peak. Carbon atoms located in the polyalcohol moiety near aromatic rings (proton peak (J) [32]) appear at higher chemical shifts (e.g., δ 64 ppm) than those closer to fatty acid chains (proton peak (I)). Unsaturated carbons of the fatty acid chains, identified as C6 in Figure 4 and assigned to the proton signal (K), exhibited chemical shifts around δ = 127 and δ = 130 ppm. Aromatic carbons attached to protons (L) and (M) (Figure 1) appeared at δ = 131 and δ = 129 ppm, respectively, which was specified using Spyros [29] as well.

GPC Analysis
The molecular weight averages, M w (weight-average molar mass) and M n (numberaverage molar mass), and dispersity (Ð) of alkyds are summarized in Table 3. With regard to Sacha inchi-based resins, M n and M w varied in the order (F)AS3 > (F)AS2 > (F)AS1, whereas, linseed-based resins did not show a specific trend. Linseed oil-based resins had the lowest M w values, which tend to decrease as the PE content increase. Linseed fatty acid-based alkyds prepared with pure glycerol presented the highest molecular weight averages values. The molecular weight distribution was correlated with the viscosity of resins, as previously reported by our research group [16,17]. As seen in Figure 7a, oil-based resins that had Gardner viscosities in the range between Z5 and Z8 presented a M w lower than 1.4 × 10 5 g/mol. High molecular weights have been related to high degrees of crosslinking [41]. On the other hand, fatty acid-based resins, despite having a greater viscosity range from Z3 to Z10, presented higher and homogeneous degrees of cross-linking. It is important to remark that fatty acid-based resins had a M w near or even higher than 10 5 g/mol. The molecular weight distribution was correlated with the viscosity of resins, as previously reported by our research group [16,17]. As seen in Figure 7a, oil-based resins that had Gardner viscosities in the range between Z5 and Z8 presented a lower than 1.4 × 10 5 g/mol. High molecular weights have been related to high degrees of cross-linking [41]. On the other hand, fatty acid-based resins, despite having a greater viscosity range from Z3 to Z10, presented higher and homogeneous degrees of cross-linking. It is important to remark that fatty acid-based resins had a near or even higher than 10 5 g/mol.  Table 1). (b) Dispersity versus molar content of pentaerythritol in alkyd composition. Table 3 clearly indicate that the size distribution of almost all resins is broad (Đ ≥ 2). There are some GPC studies of alkyd resins prepared only with glycerol as the main polyalcohol, which report the following dispersity values: (i) Jatropha Curcas oil-based alkyd, Đ = 1.15 [21]; (ii) Ricinodendron heudelotii oil-based alkyd, Đ = 1.5 [30]; (iii) Salvia Hispanica L. (Chia) oil-based alkyds, Đ = 1.2-4.3 [34]; (iv) rubber seed oil-based alkyd, Đ = 1.6 [35]. However, the oil chain length of these alkyd resins is different, as well as the saturation index of the oils used.

Results from
In general, dispersity values of oil-based resins decreased as the PE content increased (Figure 7b). GPC traces of PE-based alkyds tend to approximate a symmetric and narrow distribution, which would imply the formation of a more branched polymer (Figure 8) [42]. Đ of linseed oil-based alkyd resins is more dependent on the PE content than the alkyd resins from sacha inchi oil. Oil-based alkyd resins prepared with pure glycerol presented higher dispersity values, which could be attributed to side reactions, such as glyc-  Table 1). (b) Dispersity versus molar content of pentaerythritol in alkyd composition. Table 3 clearly indicate that the size distribution of almost all resins is broad (Ð ≥ 2). There are some GPC studies of alkyd resins prepared only with glycerol as the main polyalcohol, which report the following dispersity values: (i) Jatropha Curcas oil-based alkyd, Ð = 1.15 [21]; (ii) Ricinodendron heudelotii oil-based alkyd, Ð = 1.5 [30]; (iii) Salvia Hispanica L. (Chia) oil-based alkyds, Ð = 1.2-4.3 [34]; (iv) rubber seed oil-based alkyd, Ð = 1.6 [35]. However, the oil chain length of these alkyd resins is different, as well as the saturation index of the oils used.

Results from
In general, dispersity values of oil-based resins decreased as the PE content increased (Figure 7b). GPC traces of PE-based alkyds tend to approximate a symmetric and narrow distribution, which would imply the formation of a more branched polymer (Figure 8) [42]. Ð of linseed oil-based alkyd resins is more dependent on the PE content than the alkyd resins from sacha inchi oil. Oil-based alkyd resins prepared with pure glycerol presented higher dispersity values, which could be attributed to side reactions, such as glycerol oligomerization [43]. In the side reactions, not only glycerol monoesters but also diglycerol and diglycerol monoesters (sub products) react in the polycondensation process, possibly creating polymers of different chain lengths. Linseed fatty acid-based alkyds had dispersity below 3, while SIFA-based resins have Đ values over 2.5. FAL1 sample had the lowest dispersity. On the other hand, samples with a higher content of PE (e.g., FAS3, FAL3) had similar Đ, despite being synthesized with different fatty acid sources. Đ trend is different from that observed in the case of oilbased alkyd resins, possibly because of the differences in reaction times or minimal differences in the preparation of the fatty acids during the synthesis process. However, the alkyds prepared with the highest PE content had the lowest variation in the dispersity values. Linseed fatty acid-based alkyds had dispersity below 3, while SIFA-based resins have Ð values over 2.5. FAL1 sample had the lowest dispersity. On the other hand, samples with a higher content of PE (e.g., FAS3, FAL3) had similar Ð, despite being synthesized with different fatty acid sources. Ð trend is different from that observed in the case of oil-based alkyd resins, possibly because of the differences in reaction times or minimal differences in the preparation of the fatty acids during the synthesis process. However, the alkyds prepared with the highest PE content had the lowest variation in the dispersity values.

Conclusions
Alkyd resins were successfully characterized using NMR spectroscopy and GPC. Structural differences between polymers prepared with dissimilar polyols, glycerol and pentaerythritol, and diverse plant sources, sacha inchi and linseed oils, were found using the NMR spectroscopy. In the high field region of 1D NMR spectra, the peaks corresponding to terminal methyl groups, found in the δ = 0.8-1.1 ppm region, allowed us to identify the approximate amount and type of fatty acid with which the resins were prepared. On the other hand, 13 C NMR resonances helped to identify the type of polyalcohol that were used in the synthesis of alkyd resins, by coupling carbon atoms with oxygen atoms in the δ = 40-70 ppm region.
The 2D NMR spectroscopy facilitated the interpretation of chemical resonance assignments of the resin components. The HMQC 2D NMR spectroscopy was used for the identification of 13 C/ 1 H resonance couples, especially for the identification of methylene carbons/proton cross correlations of the fatty acid chain moieties of sacha inchi oil-based alkyd resins. With this, we can compare the degree of unsaturation of the fatty acids chains present in the alkyd resins, enabling the prediction of which resin would have better film performance properties (e.g., drying, hardness, chemical resistance).
GPC results revealed that the molecular weight averages of sacha inchi-based resins increase considerably with more PE content. On the contrary, resins prepared with linseed oil generally decreased their weight-average molar mass by containing more PE. GPC analysis confirmed that dispersity of almost all alkyds was broad. The addition of PE reduced the dispersity values of resins possibly due to the high branching generated by the polyalcohol. Alkyds based on pure glycerol may have broader molecular weight distributions because of the possible formation of side reactions during the monoglyceride process. It was also corroborated that the fatty acid manufacturing process allows synthesizing alkyd resins with a more homogeneous molecular distribution, regardless of their viscosity. Alkyd resins with similar chain lengths can be obtained by working with PE.  Data Availability Statement: The raw/processed data required to reproduce these findings cannot be shared at this time as the data also forms part of an ongoing study.