Pressure-sensitive adhesives based on soybean fatty acids

Abstract

A new class of renewable pressure-sensitive adhesive (PSA) designed and developed from soybean oil was reported in this study. Soybean oil was epoxidized and hydrolysed selectively on the ester groups to afford a mixture of epoxidized fatty acids (EFAs) which were characterized by FTIR and ¹H NMR spectroscopy. The EFA mixture without further purification was then polymerized directly in the presence and absence of a small amount of dicarboxylic acid compounds to afford hydroxyl-functionalized polymers. The peel strength, loop tack, shear strength and viscoelastic properties of the resulting (co)polymers revealed that the (co)polymers were suitable for PSA applications. The new PSAs could be fully bio-based and potentially biodegradable, and their preparation and application did not require the use of an organic solvent or a toxic chemical, thus being environmentally friendly.

---

Introduction

Pressure-sensitive adhesives (PSAs) are a distinct category of adhesives that are in a dry, solvent-free form, are aggressively and permanently tacky, and can firmly adhere to a variety of surfaces upon mere contact without the need of more than finger or hand pressure. They have sufficient cohesion strength to be cleanly removed from the adherend.¹ PSAs are viscoelastic materials with a combination of liquid-like viscous and solid-like elastic properties. Their viscous properties allow them to flow and quickly wet the adherends (even those with a rough surface) to form and develop a bond under light pressure, and their elastic properties allow them to have strong cohesive strength so that they can resist stress during the debonding stage and be cleanly removed from the adherends.^1,2 PSA products are virtually ubiquitous in homes and workplaces. Commercially available PSAs are typically based on elastomers compounded with suitable tackifiers, plasticizers and waxes. To date, the most commonly used elastomers are petrochemical-based polymers such as acrylic copolymers, styrene-based block copolymers and ethylene–vinyl acetate copolymers. Tackifiers such as C-5 and C-9 resins and waxes are also typically derived from petroleum streams.³ Petroleum is non-renewable and thus not sustainable. Furthermore, most petrochemical-based polymers are not biodegradable, thus potentially producing environmental pollution. Therefore, it is desirable to develop PSAs from renewable natural materials such as plant oils.

Plant oils are one of the most abundant and least expensive renewable raw materials that are mainly mixtures of triglycerides with various long-chain saturated and unsaturated fatty acids, depending on the plant, the crop, and the growing conditions.⁴ The triglycerides and their derivatives can be functionalized through a variety of modification reactions.^5,6 One of the most commonly used modification reactions is carried out at the double bonds of unsaturated fatty acid derivatives,⁴ e.g., the double bonds are converted into reactive epoxy groups under appropriate reaction conditions.⁵ Therefore, plant oils are one of the most attractive renewable resources and have been widely utilized as raw materials in applications such as rubbers, plastics, coatings, paint formulations, resins, and various thermosetting composites.^4,7,8 Plant oils or the fatty acid derivatives as starting materials or building blocks have numerous unique features such as renewability, low cost, low toxicity, biodegradability, and fairly high purity. In addition, the long aliphatic chains of plant oils or fatty acids impart unique properties to the resulting polymers such as elasticity, flexibility, ductility, high impact strength, hydrolytic stability, hydrophobicity, internally plasticizing effect and intrinsically low glass transition temperature.^4,9,10

However, a relatively few research efforts have been made to develop PSAs from plant oils, which can be summarized as the following general approaches. In the first approach, plant oils or fatty acid esters were first epoxidized and (meth)acrylated; the resulting monomers were then free-radically polymerized to afford PSAs. Comonomers such as 1,4-butanediol diacrylate and methacrylate were needed for the improvement of the PSA performance.^11–15 Obviously, petrochemicals were still considerably used in this approach. In the second approach, plant oils and fatty acid esters were epoxidized and polymerized to form PSAs via cationically catalyzed ring-opening polymerization; some petrochemicals such as epoxy compounds and vinyl ethers were also used as comonomers in the polymerizations.¹¹ The third approach involved the direct polymerization of the plant oils or fatty acid derivatives via the double bonds or copolymerization of them with petrochemicals such as (meth)acrylic monomers.¹¹ In this approach, however, the plant oils had to be modified to form relatively high reactive conjugated double bonds prior to the free radical polymerization. The second and third approaches are mentioned in a patent application and only theoretically work because no experimental examples of making the PSAs and no adhesive properties of the proposed PSAs are provided. Recently, a PSA was prepared by reacting epoxidized soybean oil (ESO) with phosphoric acid and formulating the resulting polymer with dihydroxyl soybean oil which was derived from ESO via a time-consuming synthetic procedure and added as a tackifier.^16–18 Dimer fatty acid-based polyesters were also reported as a viable alternative for PSAs.^19,20 However, the above procedures were time-consuming and/or considerable amount of petrochemicals and organic solvents were employed in the polymerization and/or preparation processes.

In this study, we report a novel and simple approach for development of PSAs from soybean oil. Epoxidized soybean oil was selectively hydrolysed on the ester groups to afford a mixture of epoxidized fatty acids (EFAs) that were mainly composed of AB₂- and AB-type monomers that are single molecules containing both carboxylic acid (A) and epoxy (B) groups. The EFAs mixture was then polymerized directly without further purification in the presence and absence of a small amount of dicarboxylic acids or anhydrides such as dimer acid to afford hydroxyl-functionalized polymers for PSA applications. Adhesive properties of the resulting PSAs were thoroughly investigated.

Experimental section

Materials

Epoxidized soybean oil (ESO, epoxy oxygen content, ∼7.0 wt%; iodine value, 2.0) was purchased from Spectrum Chemical Manufacturing Corp. (New Brunswick, NJ) and used as received. NaOH (97+%) and NaCl (99+%) were purchased from Acros Organics (Morris, New Jersey) and Alfa Aesar (Ward Hill, MA), respectively, and used as received. Ethanol (99.5+%), HCl (∼37% solution in water), chromium(III) tris(acetylacetonate) (97%), dimer acid (DA) (hydrogenated, average M_n ∼ 570, 98+%), adipic acid (99%), itaconic acid (99%), succinic anhydride (99%), sebacic acid (98%), phthalic anhydride (99+%), and terephthalic acid (98%) were purchased from Sigma-Aldrich Corp. (St. Louis, MO) and used as received. AMC-2 (a solution of 50% trivalent organic chromium complexes in phthalate esters) was kindly provided by Ampac Fine Chemical LLC (Rancho Cordova, CA) and used as received. The release paper (silicone coated paper) was obtained from ChemInstruments, Inc. (Fairfield, OH). Fasson 54# Semi-Gloss Facestock (paper backing, ∼68 μm thick) and the release film (polyethylene terephthalate (PET)-based) were obtained from Avery Dennison Corp. (Pasadena, CA).

Analyses

¹H NMR spectra were recorded at room temperature (∼23 °C) using a Bruker DPX400 spectrometer (Rheinstetten, Germany) operating at 400 MHz; the sample solutions were prepared in CDCl₃ (99.8% D, Cambridge Isotope Laboratories, Inc., Andover, MA). FTIR spectra were recorded on a Thermo Nicolet Nexus 470 FTIR spectrometer (Thermo Nicolet Corp., Madison, WI) coupled with Smart Golden Gate diamond ATR accessory in a scanning range of 650–4000 cm⁻¹ for 64 scans at a spectral resolution of 2 cm⁻¹. The viscoelastic property of the samples were measured with an AR 2000ex rheometer (TA Instruments, New Castle, DE) using 8 mm parallel plate geometry with an initial gap of about 1.5 mm under nitrogen atmosphere. The test samples were prepared as a film with the aid of a release paper and release PET film. The pre-polymer of the EFAs mixture as described in the following section was first coated onto a sheet of release paper with a HLCL-1000 hot-melt coater/laminator (ChemInstruments, Inc., Fairfield, OH) with the gap between the bars set at 1.5 mm. The resin side of the coated paper was then closely mated with a sheet of a release film, resulting in a “sandwich”-like assembly. The pre-polymer resin between the two release liners was cured in an air-circulating oven maintained at 160 °C for 15 min. Removal of the release liners afforded a thick film with about 1.5 mm in thickness, which was then sandwiched between parallel plates for an oscillating shear test. The strain amplitude was set at 0.1% (within the linear viscoelastic regime). Frequency sweep measurements were performed at room temperature (∼23 °C) from 0.1 to 10 Hz; temperature sweep measurements were performed at 1 Hz from −55 to 100 °C at 5 °C per step with a 40 s equilibration delay at each step.

Preparation of the EFAs mixture by the selective hydrolysis of ESO on ester groups

NaOH (10.76 g, 0.2609 mol), deionized water (48.7 mL), and ethanol (146.0 mL) were placed in a 500 mL three-necked round-bottom flask equipped with a stirrer, condenser, and thermometer. The mixture was stirred vigorously at room temperature for about 5 min to give a homogeneous solution followed by the addition of ESO (71.86 g, containing ∼0.22 mol of ester groups). The resulting mixture was then stirred at 45 °C until FTIR spectroscopy showed the complete disappearance of the ester groups (it took about 23 min) to give a yellow and slightly cloudy solution. After the solution was cooled to room temperature, vacuum was applied to the flask for 30 min to remove the majority of ethanol, resulting in an off-white “soap”. The soap was crushed to pieces and admixed with 40 mL of deionized water; the resulting mixture was then heated to 45 °C with stirring until it became homogeneous. Into the solution, 115 mL of 2.1 N HCl solution was added from an addition funnel over a period of 2 min, and the resulting milky mixture was mixed at 45 °C for another 2–3 min before being transferred to a separatory funnel that was placed in a water bath at 55–60 °C. After standing for about 5 min for phase separation, the upper organic layer was collected and washed with hot deionized water (pre-heated to 55–60 °C) for 4–5 times. Finally, the upper organic layer was washed once with hot brine (pre-heated to ∼56 °C), and collected as a white, opaque fluid which easily solidified on cooling. The product was dried under vacuum at room temperature to a constant weight of 58.0 g (ca. 85% of theoretical yield).

FTIR (neat, in cm⁻¹): 3049 and 2985 (C–H stretching of epoxy ring²¹), 2944 and 2871 (CH₃ stretching), 2911 and 2849 (CH₂ stretching), 1693 (carbonyl stretching of a carboxylic acid group (–COOH)), 1469 (CH₂ bending), 1300, 1262, 1226 and 1195 (epoxy ring symmetrical stretching, or “ring breathing”^21,22), 934 (presumably due to O–H of –COOH (ref. 22), 898 (C–C asymmetrical stretching of epoxy ring²¹), 847, 834 and 818 (“12 micron band”, typical of epoxy ring^21,22), and 719 (CH₂ rocking motions, characteristic for at least four linearly connected CH₂ groups). ¹H NMR (400 MHz, CDCl₃; δ, in ppm): 2.9–3.2 (methine protons of the epoxy groups), 2.28–2.43 (α-CH₂ adjacent to –COOH), 1.72–1.90 (CH₂ protons between the epoxy groups), 1.60–1.72 (CH₂ β to –COOH), 1.45–1.60 (CH₂ α to the epoxy groups), 1.20–1.45 (protons of other methylene groups), 0.85–1.12 (protons of methyl groups). The carboxylic acid equivalent weight (CEW) of the product was determined by titration with alcoholic KOH to phenolphthalein end point, and found to be about 300. Based on the CEW value and the ratio of the peak area of the epoxy protons at 2.9–3.2 ppm to that of the CH₂ α to carbonyl at 2.3–2.4 ppm in the ¹H NMR spectrum, the epoxy equivalent weight of the product was determined to be about 230.

Pre-polymerization

(i) Pre-polymerization of the EFAs mixture. AMC-2 (0.268 g) and the EFAs mixture (13.358 g) were placed in a 50 mL, round-bottom flask equipped with a silicon oil bath and a magnetic stirrer. The resulting mixture was bubbled with nitrogen for 2 min while stirring at 300 rpm and heating to 80 °C. The reaction mixture was stirred at 80 °C for 53 min to give a light green, homogeneous, and viscous resin. FTIR (neat, in cm⁻¹): 3445 cm⁻¹ (OH stretching), 2952 and 2871 (CH₃ stretching), 2923 and 2853 (CH₂ stretching), 1730 (ester carbonyl stretching), 1711 (–COOH carbonyl stretching), 1461 and 1378 (CH₂ bending), 1246 (ester C–O stretching), 1173 and 1112 (ester C–O–C stretching), 1073 and 1023 (C–OH stretching), 842 and 823 (“12 micron band” of epoxy ring^21,22).

(ii) Pre-polymerization of the EFAs mixture with DA. The EFAs mixture (8.544 g), DA (1.24 g), and AMC-2 (0.192 g) were placed in a 50 mL, round-bottom flask equipped with a silicon oil bath and a magnetic stirrer. The remaining procedure was the same as described previously except that the polymerization time was 50 min. FTIR (neat, in cm⁻¹): 3446 cm⁻¹ (OH stretching), 2952 and 2870 (CH₃ stretching), 2923 and 2853 (CH₂ stretching), 1733 (ester carbonyl stretching), 1709 (–COOH carbonyl stretching), 1461 and 1378 (CH₂ bending), 1246 (ester C–O stretching), 1173 and 1112 (ester C–O–C stretching), 1075 and 1025 (C–OH stretching), 842 and 823 (“12 micron band” of epoxy ring^21,22).

Curing of the pre-polymers and preparation of PSAs

Either pre-polymer resin as prepared previously was coated onto a sheet of release paper in a coating thickness of 4.0 ± 0.5 mg cm⁻² with the HLCL-1000 hot-melt coater/laminator. The resin side of the coated paper was then closely mated with a sheet of a release film with the laminator, resulting in a “sandwich”-like assembly. The pre-polymer resin between the two release liners was subjected to further polymerization and curing reaction in an air-circulating oven maintained at 160 °C. After 15 min, the assembly was taken out of the oven. The release film was peeled off without taking away any adhesive composition (the release film has better repellence against the adhesive than the release paper), and a sheet of paper backing (∼68 μm thick) was closely mated onto the adhesive layer with the laminator to give a PSA on the paper backing with the release paper left for the protection of the PSA.

Measurements of the peel strength, loop tack and shear strength of the PSAs

The peel, loop tack, and shear tests were performed with an Instron Testing Machine (model 5582, Instron, Norwood, MA).

(i) Peel test. The 90° peel strength test was performed on a stainless steel panel (type 302) in accordance with Test Method F of ASTM D3330/D3330M-04 (Reapproved 2010), with the following test parameters. The speed of the grip was 30 cm min⁻¹; the width of test specimens was 25 mm; tests were performed at 23 ± 1 °C and 50 ± 5% RH. The PSA sample was pressed with a 2 kg roller two times and allowed to wait for 1 min before each test. The strength data were collected after 25 mm of a specimen tape was peeled, and average peel strength in Newton was obtained for peeling the rest of the tape. Five specimens were tested for each sample and the averaged values were reported.

(ii) Loop tack test. The loop tack test was performed on a stainless steel panel (Type 302) in accordance with Test Method A of ASTM D6195-03 (Reapproved 2011), the maximum force required to break the adhesive bond was recorded for each specimen. The following parameters were used in the test. The crosshead speed was 30 cm min⁻¹; the contact area of PSA specimen was 25 × 25 mm²; tests were performed at 23 ± 1 °C and 50 ± 5% RH; five specimens were tested for each sample and the averaged values were reported.

(iii) Shear test. Shear tests were performed on a stainless steel panel (Type 302, mirror polished) in accordance with the Procedure A of ASTM D3654/D3654M-06. The time to failure (i.e., the tape separates completely from the test panel) was determined as the indication of shear strength. The following parameters were used in the test. Tests were performed at 23 ± 1 °C and 50 ± 5% RH; the specimen size was 25 mm by 25 mm; the test stand held the test panel (and the PSA articles applied) at an angle of about 1° with the vertical; the test mass was 1000 grams. The PSA sample was pressed with a 2 kg roller two times and allowed to wait for 1 min before each test. Five specimens were tested for each sample and the averaged values were reported.

Statistical analysis

Peel strength, tack and shear strength data were statistically analysed with a two-sample t-test using Microsoft Excel (version 2010, Microsoft Corp., Redmond, WA). All comparisons were based on a 95% confidence level.

Results and discussion

Preparation and characterization of epoxidized fatty acids (EFAs) mixture

It has been demonstrated that soybean oil can be efficiently and inexpensively epoxidized with the conversion rate of up to 98%.¹⁶ Characterization of ESO with ¹H NMR (Fig. 1, bottom) revealed that the molar ratio of the epoxy to ester carbonyl groups (epoxy/COOR ratio) was about 1.4 based on the ratio of the peak area of the epoxy protons at 2.9–3.2 ppm to that of the CH₂ α to carbonyl at 2.3–2.4 ppm in the ¹H NMR spectrum (Fig. 1, bottom).^23,24

Fig. 1 ¹H NMR spectra of ESO (bottom) and the EFAs mixture (top), with close-up views of the characteristic absorption signals related to the epoxy, –COOH, and –CH₂ α to the carbonyl groups.

ESO was readily hydrolyzed to form a mixture of EFAs with the overall yield of about 85% (Fig. 2). Characterization of the EFAs mixture with ¹H NMR (Fig. 1, top) and FTIR (Fig. 3, bottom) revealed that the ester groups were selectively hydrolysed, i.e., the epoxy groups were virtually intact during the hydrolysis. More specifically, characteristic peaks of CH₂ (4.1–4.4 ppm) and CH (5.2–5.3 ppm) of the glycerol group in ESO disappeared in the EFAs mixture, and a new broad peak of the carboxylic acid (–COOH) group proton at 6.5 ppm appeared in the EFAs mixture in the ¹H NMR spectra (Fig. 1). The FTIR spectrum of the EFAs mixture also revealed that the carbonyl peak for an ester group at ∼1740 cm⁻¹ (Fig. S1†) disappeared, and a new –COOH carbonyl peak at 1693 cm⁻¹ and a new OH peak of the –COOH group at 934 cm⁻¹ appeared.²² The characteristic three well-resolved peaks of the epoxy groups at 820–850 cm⁻¹ could be clearly observed in the FTIR spectrum (Fig. 3, bottom). In addition, diol by-products that might be generated from the ring-opening of the epoxy groups with water in the presence of NaOH were not present in the EFAs mixture, evidenced by no absorptions at 3200–3600 cm⁻¹ in the FTIR spectrum (Fig. 3, bottom) and no absorptions at 3.2–4.8 ppm (ref. 16) in the ¹H NMR spectrum (Fig. 1, top). Based on the ratio of the peak area of the epoxy protons at 2.9–3.2 ppm to that of the CH₂ α to carbonyl at 2.3–2.4 ppm in the ¹H NMR spectrum (Fig. 1, top), the epoxy/COOH ratio for the EFAs mixture was about 1.3 that was very close to the epoxy/COOR ratio (1.4) for the starting ESO.

Fig. 2 Preparation and structure of the EFAs mixture from ESO.

Fig. 3 FTIR spectra of the EFAs mixture (bottom), the pre-polymer (middle) from polymerization of the EFAs mixture in the presence of AMC-2 (∼1.6 wt% relative to the EFAs mixture) at 80 °C for 53 min, and the cured poly(EFAs) (top) from curing of the pre-polymer at 160 °C for 15 min.

In the ¹H NMR spectrum of the EFAs mixture (Fig. 1, top), distinct peaks of the methine protons of the cis-epoxy groups from epoxidized oleic acid (EOA, cis-9,10-epoxystearic acid) at 2.9 ppm (ref. 23) and those from linoleic acid di-epoxide (LADE, cis-9,10-cis-12,13-diepoxystearic acid) at 3.0 and ∼3.1 ppm (ref. 23) could be clearly identified. Unfortunately, the characteristic peaks of the methine protons of the epoxy groups from epoxidized linolenic acids could not be discernibly identified because they were supposed to appear around 2.8–3.1 ppm,²³ but were hidden inside the strong peaks associated with LADE and EOA. Because epoxidization of linolenic acid esters is not 100%,²³ epoxidized linolenic acids are expected to be a mixture of mono-, di-, and tri-epoxides. The content of these types of epoxides in the ESO and the EFAs mixture is expected to be fairly small because the linolenic acid content in soybean oil is only about 6% of all fatty acids.²⁵ The small amounts of mono-epoxides from epoxidized linolenic acids are treated as a part of EOA (a mono-epoxide), and the small amounts of the di-epoxides and tri-epoxides from epoxidized linolenic acids are treated as a part of LADE (an epoxide with more than one epoxy group). The molar ratio of LADE/EOA/other fatty acids without any epoxy groups was 54/22/24, based on the estimation from the ¹H NMR data (see the ESI† for the estimation method), which was close to that (60/21/19) for ESO.

Polymerization of the EFAs mixture and the curing of the polymers for PSAs

LADE and EOA in the EFAs mixture are AB₂- and AB-type monomers, respectively, containing both –COOH (A) and epoxy (B) groups. Theoretically, the EFAs mixture would step-growth polymerize to form hydroxyl-functionalized polyesters via the ring-opening of the epoxy groups with the –COOH groups.

A few efforts on the thermal polymerization of EOA in the absence of any catalyst have been previously reported.^26,27 However, the polymerization in the absence of any catalyst took several days to reach a satisfactory conversion of EOA, and no study on applications of the resulting polymers was reported. In addition, there was no study dealing with the polymerization of LADE or the EFAs mixture, or the potential applications of their polymers. As expected, the polymerization of the EFAs mixture at 140 °C for 10 h in the absence of a catalyst gave only a cloudy and sticky liquid with a medium viscosity. For facilitating the polymerization, various catalysts such as organometallic compounds (chromium(III) tris(acetylacetonate) (CTAA) and AMC-2), tertiary amines (N,N-dimethylbenzylamine (DMBA) and N-methylmorpholine (MMP)), and a phosphonium salt (tetraphenylphosphonium bromide (TPPB)) that have been demonstrated as effective catalysts for facilitating the reaction between –COOH and an epoxy group,^13,28,29 were then investigated. Polymerizations of the EFAs mixture in the presence of DMBA, MMP or TPPB under conditions similar to that without a catalyst afforded slightly more viscous resins than that without a catalyst, which implied that these three catalysts were not very effective in facilitating the polymerizations. In contrast, AMC-2 catalyzed the polymerization effectively and efficiently, resulting in an almost clear, light green and tacky gel in about 30 min. Interestingly, CTAA that is also a chromium(III)-based organometallic catalyst was less effective than AMC-2, requiring about 60 min instead of 30 min for the formation of the similar gel. Therefore, AMC-2 was used as a catalyst for the (co)polymerization of the EFAs mixture in this study.

The gel obtained from the polymerization of the EFAs mixture in the presence of AMC-2 was infusible and could only be swollen and partially dissolved in solvents such as chloroform, tetrahydrofuran, toluene, ethyl acetate, acetone, ethyl ether, N,N-dimethylformamide, and hexane (see the ESI† for the details), which suggested the cross-linking nature of the gel. It was difficult to evaluate adhesive properties of such a gel. Therefore, a special approach was developed for preparation of the polymeric product of the EFAs mixture into a film so that the product could be characterized and evaluated for its adhesive properties. The EFAs mixture gelled at 140 °C in the presence of AMC-2 in about 30 min, but remained spreadable at 80 °C for up to 16 h. The EFAs mixture was first polymerized at 80 °C for 53 min and then coated on a sheet of release paper to give a uniform layer at a coating weight of ca. 4.0 mg cm⁻². The coating was closely mated with a sheet of release film to afford a “sandwich”-like assembly. The composition in the assembly was then heated in an oven at 160 °C for further polymerization and curing. After completion of the curing reaction, removal of the release liners afforded a clear, pale green, uniform, flexible, and tacky film. The tacky film (Table 1, entry 1) could be easily transferred to a backing material such as paper and plastic films to form PSA labels or tapes. In this study, paper with a thickness of ∼68 μm was used as a backing material. The PSA on the paper backing was evaluated for its 90° peel strength, loop tack, and shear strength. The PSA exhibited a good peel strength (3.0 N cm⁻¹) and good tack (8 N) (Table 1, entry 1). There were no adhesive residues remained on the stainless steel panel after the peel and tack tests, which indicated that the PSA had the desirable property of adhesive failure for any commercial applications. The shear strength of the PSA had a big variation, ranging from about 28 to 135 min. Increasing the curing time from 15 min to 75 min (Table 1, entry 2) resulted in little change in the peel strength, loop tack and shear strength of the PSA. Based on these results, it is concluded that the film derived from the EFAs mixture was a suitable material for PSA applications.

Table 1 (Co)polymerization of the EFAs mixture in the presence of AMC-2, and the PSA properties of the cured polymeric products^a

Entry	Pre-polymerization^b			Cure time^d (min)	Peel strength^e (N cm^−1)		Loop tack^e (N)		Shear strength^e (min)
	Dicarboxylic acids^c (amount)	Temp(°C)	Time (min)		Average	SD	Average	SD	Average	SD
a AMC-2, 1.6 wt% relative to the reaction mixture; PSA layer thickness, 4.0 ± 0.5 mg cm^−2; paper (∼68 μm thick) was used as the backing material.b The pre-polymerization took place in the flask with stirring.c Molar percent relative to the epoxy groups of the EFAs mixture.d For curing, the pre-polymer was sandwiched between release liners and placed in an oven at 160 °C.e See the Experimental section for the test methods applied; the failure mode in peel and loop tack tests was adhesive failure; SD, standard deviation, five specimens were tested for each adhesive; n.m., not measured.
1	—	80	53	15	3.0	0.5	8.0	1.5	55	45
2	—	80	53	75	2.8	0.5	7.3	1.6	n.m.
3	DA (6.0)	80	50	8	2.5	0.4	7.5	1.5	100	90
4	DA (12.0)	80	50	8	3.6	0.6	8.9	1.7	n.m.
5	SA (11.9)	100	25	7	3.4	0.6	8.8	1.5	n.m.
6	AA (12.5)	100	25	5	3.3	0.5	8.0	1.8	n.m.
7	IA (12.6)	100	25	5	2.1	0.3	n.m.		n.m
8	SAn (13.0)	120	5	4	1.3	0.2	n.m.		n.m.
9	PAn (13.0)	120	10	10	1.0	0.3	2.6	0.5	420	170
10	TPA (11.0)	120	22	11	1.5	0.3	3.6	0.8	350	110

---

For understanding the polymerization and curing, the whole process of making the PSA from the EFAs mixture was monitored by FTIR spectroscopy. As shown in the FTIR spectra (Fig. 3 and S2†), a new broad peak of the hydroxyl group at 3450 cm⁻¹ and a new peak of the ester group at 1732 cm⁻¹ developed during the polymerization, with the concomitant decrease of the peak of the epoxy groups between 820–850 cm⁻¹ and that of –COOH at ∼1700 cm⁻¹ over time, which confirmed that the ring-opening reactions of the epoxy groups with the –COOH groups took place, producing new hydroxyl groups and new ester linkages under the experimental conditions used in this study. After curing at 160 °C for 15 min, the epoxy groups almost disappeared, while a small amount of –COOH groups was still present as evidenced by a slight shoulder on the right side of the ester carbonyl peak in the FTIR spectrum (Fig. 3, top). Increasing the curing time from 15 min to 75 min did not change the intensities of peaks of the unreacted –COOH groups, the remaining epoxy groups and other functional groups (Fig. S2†), which implied the reactions were complete after 15 min.

The monitoring of the polymerization of the EFAs mixture with FTIR also revealed that the epoxy groups were consumed faster than the –COOH groups throughout the polymerization, which implied that except their reactions with –COOH groups, the epoxy groups also experienced other side reactions. Possible side reactions might include those between the epoxy groups and secondary hydroxyl groups or water (hydrolysis), and those between the epoxy groups themselves.²⁷ The faster consumption of epoxy groups than –COOH groups was also observed in the polymerization of EOA.^26,27 As the polymerization progressed, the amount of –COOH groups decreased, and the diffusion of the remaining end-capping –COOH became more and more restricted because the molecular weight of the resulting polymers and the viscosity of the reaction mixture increased. In the meantime, the amount of hydroxyl groups increased along with the polymerization reactions. Therefore, the reactions between the epoxy and –COOH groups decreased and side reactions increased as the polymerization reactions approached to the end. Ether linkages from the ring-opening reaction of the epoxy groups in epoxidized fatty acid esters have been reported to have a characteristic peak at ∼1075 cm⁻¹ in the FTIR spectrum.^16,30 A peak at ∼1073 cm⁻¹ in the FTIR spectrum of the polymers increased in intensity during the polymerization of the EFAs mixture (Fig. 3, middle vs. top), which might indicate the formation of ether linkages. This evidence is not conclusive though because the content of ether linkages is much lower than the ester and hydroxyl groups that also have peaks in 1000–1200 cm⁻¹ region due to C–O and C–OH stretching, respectively. Formation of ether linkages in the polymerization of EOA was also observed and confirmed by ¹H NMR analysis (spectrum not shown) in our lab.

The EFAs mixture contained about 54 mol% of LADE that contained two epoxy groups per fatty acid chain. After the first epoxy group of LADE reacted with a –COOH group to form an oligomeric/polymeric chain, the second epoxy group could react with free, i.e., unreacted fatty acids to form branches (Fig. S3,† top). If the free fatty acids did not contain an epoxy group, the branches became dead ends. If the free fatty acids contained an epoxy group, the branches could lead to formation of new oligomeric/polymeric chains that could lead to more branches. At the beginning of the polymerization of the EFAs mixture at a low temperature (80 °C), the mixture step-growth polymerized primarily via the epoxy–COOH reaction to afford hydroxyl-functionalized, branched or highly branched oligomers. Further heating at the same temperature or an elevated temperature (160 °C) led to further step-growth polymerization and the augmentation of the molecular weight and branches of the polymers (Fig. S3,† bottom). When a critical conversion of the epoxy and –COOH groups was reached, the reaction mixture turned highly viscous, and further heating at 160 °C led to cross-linked polymer networks (Fig. 4, Route A). Although the epoxy–COOH reaction was the major reaction, other side reactions such as those of the epoxy groups reacting with secondary hydroxyl groups to give ether linkages (Fig. 4, Route A) would be enhanced toward the end of the polymerization when more hydroxyl groups were generated and the reaction mixture became more viscous. These side reactions contributed to the cross-linking of the polymers and accounted for the gelation.

Fig. 4 A schematic representation of possible polymeric networks resulted from the polymerization of the EFAs mixture in the absence (Route A) and presence of a dicarboxylic acid (Route B), and of typical functional groups, branching and cross-linking sites.

Copolymerization of the EFAs mixture with a dicarboxylic acid

The EFAs mixture had the epoxy/COOH molar ratio of ∼1.3/1, thus being stoichiometrically imbalanced between the epoxy and –COOH groups. A dicarboxylic acids or dicarboxylic acid anhydride was added to the EFAs mixture for adjusting the epoxy/COOH molar ratio to 1/1, hoping for increasing the epoxy–COOH reactions and minimizing side reactions. The following dicarboxylic acids or anhydrides with different backbone flexibility and/or length of the spacer groups between the –COOH groups were investigated: dimer acid (hydrogenated, DA), sebacic acid (SA), adipic acid (AA), itaconic acid (IA), terephthalic acid (TPA), succinic anhydride (SAn), and phthalic anhydride (PAn). The resulting adjusted mixtures were polymerized under otherwise the same conditions as those used in the polymerization of the EFAs mixture only, with the experimental conditions and results summarized in Table 1.

DA is a long-chain dicarboxylic acid derived from unsaturated fatty acids, and has been widely used in coatings and adhesives because of the unique properties it imparts to the resulting polymers such as elasticity, flexibility, ductility, high impact strength, hydrolytic stability, hydrophobicity, and intrinsically low glass transition temperature (T_g).³¹ Copolymerization of the EFAs mixture with 6.0 mol% of DA (relative to the epoxy groups; Table 1, entry 3) and the subsequent curing of the polymers between the release liners afforded a clear, pale green, uniform, flexible, elastic, and tacky film that had enough cohesive strength not to leave any adhesive residues on the stainless steel test panel during the 90 degree peel test. The curing reactions were complete after 8 min (Fig. S4†) which was shorter than that (15 min) for the EFAs mixture only (Table 1, entry 1). The increase of the –COOH groups in the mixture increased the rate of the epoxy–COOH reaction, thus speeding up the polymerization and curing which were primarily based on the epoxy–COOH reaction. The resulting PSA (Table 1, entry 3) exhibited a peel strength slightly lower than that from the polymerization of the EFAs mixture only, which could be due to an increase in the cross-linking density of the resulting copolymer.¹² The EFAs–DA mixture had the epoxy/COOH molar ratio of ∼1.1/1, which was almost stoichiometrically balanced between the epoxy and –COOH groups. Under this condition, the mixture could most probably result in oligomers or polymers with more than one –COOH group (Fig. S5,† middle). As a comparison, the polymerization of the EFAs mixture only afforded oligomers or polymers with only one –COOH group (Fig. S3,† bottom). The oligomers or polymers with more than one –COOH group had more chance of cross-linking between the oligomers and/or polymers via the epoxy–COOH reaction (Fig. S5,† bottom and Fig. 4, Route B) than those with only one –COOH group.

Further increasing the amount of DA to 12.0 mol% (Table 1, entry 4) did not shorten the cure time any further. However, the resulting PSA exhibited an increase in the peel strength, which could be the result of a higher amount of unreacted –COOH group remaining in the final PSA as compared to the copolymerization with a lower amount of DA (6.0 mol%). The –COOH groups could improve the wetting onto the stainless steel surface and accelerating the rate of bond establishment and development via the formation of hydrogen bonding and other noncovalent interactions.^19,20,32 Such effect of –COOH groups in acrylic-based adhesives obtained by copolymerization of acrylic esters with acrylic acid was well-documented.^33,34

As with DA, the addition of SA, AA, IA, SAn, PAn and TPA to the EFAs mixture also shortened the cure time (Table 1, entries 5–10). However, the copolymerizations (Table 1, entries 4–8) afforded PSAs exhibiting peel strength in a decreasing order: DA ∼ SA ∼ AA > IA > SAn. Speculatively, the decrease in the spacer length of the dicarboxylic acids resulted in less flexibility and higher T_g of the resulting copolymers, which could result in the decrease in the peel strength and tack.³¹ The PSAs from the copolymers of the EFAs mixture with aromatic dicarboxylic acids PAn and TPA exhibited a lower peel strength and loop tack (Table 1, entries 9 and 10), but an obviously higher shear strength (420 ± 170 min and 350 ± 110 min for the PAn and TPA systems, respectively) than that (100 ± 90 min) of the PSA obtained from the copolymerization with DA. It was reported that rigid moieties or blocks such as aromatic moieties in the polymers could play multiple roles such as adjusting the T_g and modulating the viscoelastic properties of the polymers, therefore, the resulting polymers had enhanced elastic modulus, and was more cohesive than those without any rigid moiety.^19,38 The rigid phenylene spacer from PAn and TPA could increase the stiffness of the resulting polymer chains and the T_g of the resulting copolymers, which could account for the decrease in the peel strength and loop tack.³¹ On the other hand, incorporation of phenylene moieties into the polymer backbones that are sufficiently mobile to allow ordered chain packing could result in physical cross-links in the polymer network.³⁵ These physical cross-links increased the inter-chain cohesion and the shear strength of the PSAs.¹²

Commercially available office tapes typically have a peel strength ranging from about 3.5 to 8.8 N cm⁻¹,¹³ and the requirement of the peel strength for office labels such as Post-it® note can be as low as ∼0.7 N cm⁻¹.¹⁷ The peel strength of the PSAs from this study ranged from 1.0 to 3.6 N cm⁻¹ (Table 1), which implied that they could potentially be used as office tapes and labels. The PSAs can be solely bio-based because the EFAs mixture, DA, IA and SA can be derived from renewable bio-materials.^31,36 The PSAs are very environmentally friendly products because no organic solvents or toxic chemicals are used in their preparation.

Viscoelastic properties of the PSAs obtained from (co)polymerization of the EFAs mixture

The cured polymeric mixtures from the (co)polymerization of the EFAs mixture consisted of cross-linked networks that were not soluble in the solvents described previously and dispersed sol fractions that were soluble in the solvents. Representative possible network structures with typical functional groups, branching and cross-linking sites were schematically illustrated in Fig. 4. The sol fractions could consist of hydroxyl-functionalized polymers and oligomers including those terminated and/or branched by monofunctional fatty acids (bearing –COOH but no epoxy groups), and possible loops and unreacted free fatty acids. These mixtures turned out to be suitable materials for PSAs in this study.

For gaining a good understanding of the PSA nature of the cured polymeric mixtures, a typical EFAs-based PSA obtained from the copolymerization of EFAs and DA (Table 1, entry 3) was characterized with a TA rheometer (AR 2000ex) for its viscoelastic properties (Fig. S6†). The frequency sweep measurement at 23 °C showed a dynamic storage modulus (G′) of 2.7 × 10⁴, 6.0 × 10⁴ and 1.5 × 10⁵ Pa at 0.1, 1.0 and 10 Hz, respectively, which indicated that the G′ was highly dependent upon the frequency. The strong dependence of the G′ on the frequency further indicated that the new PSA was a soft material with a low cross-linking degree.²⁰ The small amount of cross-linking proved to be sufficient to suppress the global flow of free chains so that the PSA possessed sufficient cohesion strength while maintaining good tack.²⁰ The rheological measurement showed that G′ (2.7 × 10⁴ Pa) of the PSA at room temperature and 0.1 Hz was clearly below 3 × 10⁵ Pa, which satisfied the Dahlquist criterion for tack.^2,37 According to the criterion, G′ of an adhesive must be 3 × 10⁵ Pa or smaller for effective promotion of the wetting and the maximum contact of the adhesive with the adherend and for exhibition of good tack characteristics. Furthermore, the PSA showed a loss modulus (1.6 × 10⁴ Pa) slightly lower than G′ (2.7 × 10⁴ Pa) at a low frequency of 0.1 Hz at room temperature, which was also correlated well with the tacky nature of the PSA.²⁰

It is generally agreed that the T_g of the PSAs needs to be lower than the temperature which the PSA products is intended to be used at.² At or near T_g, the polymer segment starts to move and the material has its greatest ability to loose energy through deformation; the tackiness is greatest here.^1,2 In practice, the T_g of polymers that are intended to be used in PSA applications at room temperature generally falls between −15 and −55 °C.¹⁹ For our EFAs-based PSA (Table 1, entry 3), a temperature sweep at 1 Hz revealed a low T_g of about −15 °C, which was expected, considering the inherent flexibility and intrinsically low T_g of the polymers imparted by the long-chain-length fatty acid-derived monomers.³¹ Macroscopically, the PSA was soft, flexible and tacky, and could readily deform and wet adherends, thus facilitating surface macromolecular rearrangement of the PSA layer in contact with the adherends so as to increase the interfacial energy.^1,2

In addition, hydroxyl groups (–OH) and a small amount of unreacted –COOH groups were present in our PSAs. Incorporation of such functional groups into a PSA is known to improve the adhesion strength and cohesion of a PSA.^33,34,38 In the process of bond establishment and development, the functional groups –OH and –COOH could improve the wetting by imparting good attraction to adherends such as paper, metal, glass, and skin, and play the role as adhesion promoters by the development of hydrogen bonds and other non-covalent interactions near the adhesive/adherend interface.^19,20,32 Therefore, the –OH and –COOH groups in the new PSAs contributed to the good tack and good adhesion strength. On the other hand, the presence of –OH and –COOH groups in the polymer networks and sol fractions could also contribute to the inter-chain cohesion between the networks and/or sol fractions via intermolecular hydrogen bonding interactions, which increased the cohesion strength of the PSAs.³⁹

Conclusions

The alkaline hydrolysis of epoxidized soybean oil was highly selective and resulted in an EFAs mixture with the complete retention of the epoxy group. The EFAs mixture without further purification was polymerized and cured to afford a hydroxyl-functionalized polymer. The peel strength, loop tack and shear strength of the polymer revealed that it was suitable for PSA applications. The addition of a small amount of a dicarboxylic acid or anhydride such as DA, SA, AA, IA, SAn, PAn and TPA to the EFAs mixture increased the rate of the polymerization and curing. The PSA properties of the resulting copolymers could be tailored for different applications such as tapes and labels through the selection of a dicarboxylic acid and its usage. This new class of PSAs could be fully based on renewable materials. The preparation and application of the PSAs were very environmentally friendly.

Acknowledgements

This research was supported by the royalty fee income of patented technologies from K. Li's research group.

References

1. Handbook of Pressure-Sensitive Adhesives and Products: Fundamentals of Pressure Sensitivity, ed. I. Benedek and M. M. Feldstein, Taylor & Francis Group, Boca Raton, FL, 2009, vol. 1, pp. 1–2.
2. S. G. Chu, in Handbook of Pressure Sensitive Adhesive Technology, ed. D. Satas, Van Nostrand Reinhold, New York, 2nd edn, 1989, ch. 8, pp. 158–203.
3. Handbook of Pressure-Sensitive Adhesives and Products: Technology of Pressure-Sensitive Adhesives and Products, ed. I.Benedek and M. M. Feldstein, Taylor & Francis Group, Boca Raton, FL, 2009, vol. 2, pp. 1–2.
4. M. A. Meier, J. O. Metzger and U. S. Schubert, Chem. Soc. Rev., 2007, 36, 1788–1802.
5. U. Biermann, W. Friedt, S. Lang, W. Lühs, G. Machmüller, J. O. Metzger, M. R. Klaas, H. J. Schäfer and M. P. Schneider, Angew. Chem., Int. Ed., 2000, 39, 2206–2224.
6. H. Baumann, M. Bühler, H. Fochem, F. Hirsinger, H. Zoebelein and J. Falbe, Angew. Chem., Int. Ed., 1988, 27, 41–62.
7. V. Sharma and P. P. Kundu, Prog. Polym. Sci., 2006, 31, 983–1008.
8. C. K. Williams and M. A. Hillmyer, Polym. Rev., 2008, 48, 1–10.
9. D. Ratna and A. K. Banthia, J. Adhes. Sci. Technol., 2000, 14, 15–25.
10. P. Czub, Macromol. Symp., 2006, 245–246, 533–538.
11. C. A. Koch, Int. Patent Appl. WO 2008/144703 A2, 2008.
12. S. P. Bunker, C. Staller, N. Willenbacher and R. P. Wool, Int. J. Adhes. Adhes., 2003, 23, 29–38.
13. S. P. Bunker and R. P. Wool, J. Polym. Sci., Part A: Polym. Chem., 2002, 40, 451–458.
14. C. M. Klapperich, C. L. Noack, J. D. Kaufman, L. Zhu, L. Bonnaillie and R. P. Wool, J. Biomed. Mater. Res., Part A, 2009, 91, 378–384.
15. S. B. David, K. Sathiyalekshmi and G. A. G. Raj, J. Mater. Sci.: Mater. Med., 2009, 20, S61–S70.
16. B. K. Ahn, S. Kraft, D. Wang and X. S. Sun, Biomacromolecules, 2011, 12, 1839–1843.
17. B. K. Ahn, S. Kraft and X. S. Sun, J. Mater. Chem., 2011, 21, 9498–9505.
18. B. K. Ahn, J. Sung and X. S. Sun, J. Am. Oil Chem. Soc., 2012, 89, 909–915.
19. R. Vendamme and W. Eevers, Macromolecules, 2013, 46, 3395–3405.
20. R. Vendamme, K. Olaerts, M. Gomes, M. Degens, T. Shigematsu and W. Eevers, Biomacromolecules, 2012, 13, 1933–1944.
21. R. M. Silverstein and F. X. Webster, Spectrometric Identification of Organic Compounds, John Wiley & Sons, New York, 6th edn, 1998, p. 91.
22. O. D. Shreve, M. R. Heether, H. B. Knight and D. Swern, Anal. Chem., 1951, 23, 277–282.
23. B. K. Sharma, K. M. Doll and S. Z. Erhan, Green Chem., 2007, 9, 469–474.
24. G. Du, A. Tekin, E. G. Hammond and L. K. Woo, J. Am. Oil Chem. Soc., 2004, 81, 477–480.
25. B. K. Sharma, A. Adhvaryu and S. Z. Erhan, J. Agric. Food Chem., 2006, 54, 9866–9872.
26. B. H. Nicolet and T. C. Poulter, J. Am. Chem. Soc., 1930, 52, 1186–1191.
27. D. Swern, G. N. Billen and C. R. Eddy, J. Am. Chem. Soc., 1948, 70, 1228–1235.
28. W. J. Blank, Z. A. He and M. Picci, J. Coat. Technol., 2002, 74, 33–41.
29. S. N. Khot, J. J. Lascala, E. Can, S. S. Morye, G. I. Williams, G. R. Palmese, S. H. Kusefoglu and R. P. Wool, J. Appl. Polym. Sci., 2001, 82, 703–723.
30. G. Lligadas, J. C. Ronda, M. Galia and V. Cadiz, Biomacromolecules, 2006, 7, 3521–3526.
31. K. Hill, Pure Appl. Chem., 2000, 72, 1255–1264.
32. Y. Peykova, S. Guriyanova, O. Lebedava, A. Diethert, P. Muller-Bushbaum and N. Willenbacher, Int. J. Adhes. Adhes., 2010, 30, 245–254.
33. C. Gay, Integr. Comp. Biol., 2002, 42, 1123–1126.
34. A. Bellamine, E. Degrandi, M. Gerst, R. Stark, C. Beyers and C. Creton, Macromol. Mater. Eng., 2011, 296, 31–41.
35. J. E. White, D. J. Brennan, H. C. Silvis and M. N. Mang, in Specialty Monomers and Polymers: Synthesis, Properties, and Applications, ed. K. O. Havelka and C. L. McCormick, American Chemical Society, Washington, DC, 2000, vol. 755, ch. 10, pp. 132–146.
36. U. Edlund and A. C. Albertsson, Adv. Drug Delivery Rev., 2003, 55, 585–609.
37. C. A. Dahlquist, Adhes. Age, 1959, 2, 25–29.
38. M. D. Gower and R. A. Shanks, J. Polym. Sci., Part B: Polym. Phys., 2006, 44, 1237–1252.
39. J. E. White, H. C. Silvis, M. S. Winkler, T. W. Glass and D. E. Kirkpatrick, Adv. Mater., 2000, 12, 1791–1800.

---

Footnote

† Electronic supplementary information (ESI) available: Estimation of the fatty acid distribution in ESO and the EFAs mixture, and additional FTIR spectra. See DOI: 10.1039/c4ra03557g

---