Page 1

IMPROVING BOTH PERFORMANCE AND SUSTAINABILITY OF THERMOPLASTIC POLYURETHANES (TPUs) VIA BIO-BASED

SUCCINATE-POLYESTER POLYOLS (SA-PEPs)

William D. Coggio, Ph.D. 1 BioAmber Inc.2

Dan Pseja3 and Choung Bin Kang4 Polyurethane Specialties Corp.

Alan Schrock, Ph.D.5 and Natalie Dzadek6 Department of Chemistry, University of West Florida

Abstract As an emerging bio-based chemical, bio-based succinic acid (BBSA) is produced via a

highly efficient, yeast-based fermentation process that reduces CO2 emissions and energy requirements vs. petrochemical-based organic acids (e.g. succinic acid (p-BA) or adipic acid (AA)) while also yielding a high-quality, polymerization-grade succinic acid (C4H6O4). When used as a platform chemical, BBSA can be reacted with glycols to produce succinate-polyester polyols (SA-PEPs). These monomers are useful in polyurethane (PU) and polyester chemistry. They can be used much like adipic acid-based polyester polyols (AA-PEPs), although they generally exhibit excellent strength and elongation properties with excellent solvent resistance vs. AA-PEPs and help produce thermoplastic polyurethanes (TPUs) with 60% or greater renewable carbon ― a feature that is quite desirable to many industries, including automotive. BBSAs and SA-PEPs offer chemists the formulation flexibility to produce TPUs with both higher performance and sustainability.

This paper describes recent work comparing TPUs made with different SA-PEPs and explores the effects of SA-PEP composition on viscosity, hydrolysis rate, and physical and mechanical properties of the resultant polymers. It concludes with a discussion of how these properties influence the performance range of the resulting bio-based TPUs in elastomeric applications.

1 Global Manager Applications and Technology Support, BioAmber Inc., 3850 Annapolis Lane, Plymouth, MN 55473 USA, https://www.bio-amber.com, bill.coggio@bio-amber.com. 2 Bio-Amber Disclaimer: This information and our technical advice - whether verbal, in writing or by way of trials - are given in good faith but without warranty, and this also applies where proprietary rights of third parties are involved. Our advice does not release you from the obligation to verify the information currently provided - especially that contained in our safety data and technical information sheets - and to test our products as to their suitability for the intended processes and uses. The application, use and processing of our products and the products manufactured by you on the basis of our technical advice are beyond our control and, therefore, entirely your own responsibility.

3 Technical Director - Polyurethane Specialties Corp., Lyndhurst, NJ USA, http://www.polyurethanespecialties.com/. 4 Research Specialist, Polyurethane Specialties Corp., Lyndhurst, NJ, http://www.polyurethanespecialties.com/. 5Professor and Chair, Department of Chemistry, University of West Florida, Pensacola, FL USA, www.uwf.edu/cseh/departments/chemistry. 6 Former Chemistry Student, Department of Chemistry, University of West Florida, now a Ph.D. Student in atmospheric science, University of Georgia, Athens, GA USA, http://www.uga.edu/.

Page 2

Introduction Urethane chemistry has made use of bio-based building blocks for decades. For example,

bio-based glycerin, sucrose, and sorbitol all have been used as initiators for polyether polyols. Also, natural oil-based polyols, fatty acid dimers, and diacids (e.g. sebacic acid (C10-diacid) have been used as building blocks for polyester polyols. Recent advances in fermentation have led to development of bio-based C4 succinic acid, which is a key building block for the polyester polyols used both in urethanes as well as other ether-based resin formulations.

In recent years, several papers have been published on the use and performance of BBSA and SA-PEPs in a variety of applications [1-2]. Via several commercial announcements, the performance benefits of succinate polyols in urethanes also have been substantiated in diverse application areas such as synthetic leather, thermoplastic elastomers and plastics, urethane coatings for textiles and wood, BBSA-based succinic acid ester lubricants, and personal care emollients7. Thanks to the shorter C4 diacid, some unique property trends as well as greater sustainability can be brought to these differing applications. This paper will highlight recent advances in BBSA-based polyester polyols for TPUs made by a cast-polyurethane (CPU) process.

Experimental Using a well-known commercial process, a series of SA-PEPs and AA-PEPs were prepared

by Polyurethanes Specialties Corp. (PSC, Lyndhurst, NJ, USA). Polyols with molecular weight of ≈2,000 g/mol were produced by controlling the hydroxyl number to 56± 3 mg-KOH/g. Key polyol properties are shown in Table I. The glycols used in the study were as follows:

• MPD = 2- methyl 1,3 propane diol,

• BDO = 1,4 butane diol,

• HDO = 1,6 hexane diol,

• PDO = 1,3 propane diol8.

PSC also prepared molded TPUs via a CPU process where a given polyol was pre-reacted

using 4,4'-MDI, producing an NCO-terminated pre-polymer containing 6 wt% NCO end groups. The NCO pre-polymer was then reacted with the chain extender (1,4 butane diol), whereupon it was transferred to a sheet mold and cured at 100°C for 2 hr, followed by a 4-hr post-cure at the same temperature. To minimize residual catalyst influence in the study, no additional isocyanate was used to facilitate the urethane reaction. Final sheet dimensions were ≈30.5 cm2 at a thickness of 2 mm.

7 On April 24, 2015, Bayer MaterialScience (now Covestro) announced that BioAmber was a supplier for the their new IMPRANIL™ Eco product line for bio-based urethanes. On June 23, 2015, Flokser Group announced the introduction of Sertex© Synthetic Leather utilizing BioAmber’s bio-based succinic acid and Susterra™ 1,3-propane diol for DuPont Tate and Lyle to produce a bio-based synthetic leather with about 70% bio-based content and excellent performance properties. For other press releases, see https://www.bio-amber.com/bioamber/en/news. 8 Susterra™ from DuPont Tate and Lyle Bioproducts.

Page 3

Cured sheets were conditioned at room temperature (RT) for 1 wk before being die cut into specimens for testing. As a control, samples were produced by PSC using AA-BDO and PTMEG from commercial formulations9. Next, Troy Polymers, Inc. (Troy, MI, USA) conducted mechanical tests on the specimens using previously published test methods for PU mechanical property evaluations [4-6]. Another company, Assured Testing (Ridgeway, PA, USA) conducted Taber abrasion tests according to ASTM D4060-10 using an H22 abrasive wheel and a 500-g weight. Results were recorded as milligram weight loss (mg). Last, Bashore rebound testing for resiliency was conducted by PSC.

Measuring solvent absorption

Polymer samples were prepared as follows: 2-mm thick material was cut into pieces ≈20 x 20 mm and then dried to a constant weight under vacuum. Next, cured samples were weighed on a Mettler AE200 balance to 4 decimal places. The samples were then submerged in common solvents (e.g. room-temperature water, ammonium hydroxide, ethanol, methyl ethyl ketone (MEK), toluene, and ethyl acetate, as well as IRM 903 reference oil at 63°C), which were used as received. Solvent and samples were placed in closed glass jars and held at both room and elevated temperatures (the latter in an oven). During the period of testing, samples from each solvent were removed once per day, dried with a paper towel, and weighed before being returned to the solvent. This was done for 12 days until the samples equilibrated. Duplicate samples for each formulation and test condition were run and average weight changes were recorded. A control specimen for each formulation that was exposed only to air was used to determine if significant swelling occurred due to atmospheric moisture (it did not). Each control also was used to measure the density of that formulation, with measurements being taken at the start and finish of the test.

DSC characterization

Next, differential scanning calorimetry (DSC) was used to measure the glass-transition temperature (Tg) and melting point (Tm) for the PEP and CPU samples using a Toledo DSC 1 Star System. A temperature ramp rate of 10°C/min was used and samples were heated from 30°C to 220°C, then cooled to -60°C, followed by reheating at the same rate to 220°C again. Tg and Tm values were recorded during the transition detected using the second heating, with Tg values being recorded at the midpoint of the transitions and a Tm value range reported at onset and peak melt values for the endotherm.

Calculating bio-based carbon content

Unless otherwise noted, Beta Analytic Inc. (Miami, FL, USA) calculated bio-based carbon content using C14 analysis per ASTM 6866 by estimating the total carbon count in a molecule and dividing by the bio-based contribution. To estimate the bio-based carbon content of the polyol based on adipic acid and propane diol (AA-PDO), the number of bio-based carbons (3 for PDO) was divided by the total number of carbons in the molecule (3/9 or 33.3%).

9 ME7-55D and PCA-6-3, both from PSC..

Page 4

Table I: Select properties of polyester polyols used in study (ND = not detected by DSC; NR = not reported by supplier)

MPD = 2- methyl 1,3 propane diol, BDO = 1,4 butane diol, HDO = 1,6 hexane diol, PDO = 1,3 propane diol

Polyol Abbreviation

SA-MPD

SA-MPD/BDO

SA-HDO/BDO

AA-HDO/BDO SA-PDO AA-PDO

AA-BDO PTMEG

diol Type (glycol ratio 50% mole)

2-methyl

1,3-propane

diol

2-methyl 1,3-

propane diol and

1,4 butane

diol

hexane diol and

1,4 butane

diol

hexane diol and

1,4 butane

diol

Susterra™ bio-based

1,3 Propane

diol

Susterra™ bio-based

1,3 Propane

diol

1,4 butane

diol

poly tetra-methylene

oxide

OH# (mg-KOH/g polyol)

55 55 56 54 55 53 56 NR

Average number of repeat units, n, for 2000 g/mol

11.1 11.1 10.2 8.6 12.2 10.3 9.6 NR

Polyol Viscosity (cps, 60°C) 5100 3100 2360 1400 6500 2800 1800 NR

Prepolymer Viscosity (cps, 100°C) 1760 1800 1620 1300 2160 1640 590 1420 wt%-NCO prepolymer 6.1 5.93 6.0 6.0 6.1 6.1 6.1 6.6

Tm Polyol (peak, °C) ND 45 56 30-36 ND 35 65 NR

Tg Polyol (onset °C) -37 -43 -54 -69 -38 -61 ND NR

Tg Polyol (midpoint °C) -35 -41 -52 -60 -37 -59 ND NR Calculated New Carbon Content (NCC) 50% 50% 45% 0 100% 33% 0 0

Page 5

Results & Discussion Characterizing SA-Polyol

Table I shows key physical properties that differentiate the various PEPs used to make TPUs in this study. These PEPs were prepared via standard polyol reaction equipment and conditions. No significant differences in polyol synthesis reaction rates were noted; however, a comparison of structure-property relationships between SA-PEPs and AA-PEPs found increased Tg and polyol viscosity (η) as shown in Figures 1 & 2. Researchers believe this was caused by the shorter 4-carbon succinic acid chain, which decreases the number of atoms between ester groups, thereby reducing the degree of freedom for bond rotation, which in turn leads to higher Tg and η.

Not surprising, the glycol's structure also influenced polyol properties. Researchers believe the average length of the glycol increases Tg and decreases η. Some state that if the sum of carbons in the SA-glycol combo results in an even number, then the polyols will be solid at RT, while if the sum yields an odd number, they likely will be liquid or soft waxes at 25°C [2]. Although the tendency for SA-PEPs to have higher viscosity is thought to be offset by using long-chain, branched, or odd-numbered glycols, interestingly it has been found that polyols based on glycol combinations with SA-BDO/X and containing at least 50 mol% of BDO will solidify at RT even if X is a branched glycol or one with an odd number of carbons. Effectively that means that SA-BDO/X with 2-MPD, 3-MPD, and PDO are solids at RT, while SA-BDO/DEG has no detectable melting endotherm. These and other observations are shown in Table II for a series of copolyester polyols with the general chemical Structure 1 where the ratio of R/R' is 1/1 by mole.

NCO-prepolymer and CPU preparation and characterization

The TPU polymers used in this study were readily synthesized by reacting bio-based succinate polyester polyols (2) with 4,4' methylene diphenyl diisocyanate (MDI) to form the isocyanate-terminated polyester urethane prepolymer (3), which contained ≈6 wt% NCO end groups as is shown in Scheme 1. Next, NCO prepolymer (3) was mixed at 80-100C with BDO (acting as the chain extender) and cast into a mold. The polymer was then polymerized at 100°C for 60 min, and post-cured (annealed) at the same temperature for 24 hr to ensure high conversion to the final CPU (4).

Figure 2 displays the viscosity values of the NCO prepolymer (3), which can be compared with that for the polyester polyols (2) listed in Table I. As previously noted, viscosity values for the NCO prepolymers were similar regardless of the nature of the polyol (2), ranging from ≈2,100 cps for SA-PDO to ≈1,600 cps for SA-HDO/BDO. While slightly higher in viscosity than corresponding adipate prepolymer systems, they do fall within a very usable processing range.

Page 6

Figure 1: Glass transition temperatures (Tg °C) for SA-polyester polyols (SA-PEP) and AA-polyester polyols.

Figure 2: Brookfield viscosity of polyester polyols and NCO prepolymers (see Scheme 1, Structures 2 and 3)

Page 7

Table II: Glass transition and melting temperature matrix for succinate polyester polyols (SA-PEP) with mixed glycols based on a 1:1 mole ratio

(Tg, value is shown in the top portion of each cell and Tm value is shown in lower portion of each cell. Thermal transition data are based on polyols with ~1,000 g/mol. Some values also cited from [Ref. 4 & 5].)

OO R/R' OH

O

O

R/R'

n

HO

Structure 1: General structure of series of copolyester polyols used in the study

Page 8

Preparation of TPU by the prepolymer method used here is different from the more common "TPU one-shot method" [4-6]. With the prepolymer method, a well-defined polyester soft block is formed in the PU microstructure. Scheme 1 shows that only PEP can react with isocyanate in the initial stages of the prepolymer method to form the NCO prepolymer intermediate (3). In the one-shot method, however, the PEP, BDO (chain extender), and MDI are mixed together and all compete for isocyanate to form polyester soft blocks, which contain both BDO chain extender and polyester polyol in the microstructure. This difference means that polymers with dissimilar microstructures could inadvertently be prepared using the one-shot method, and researchers feel that could influence end-use properties of the resultant TPU10.

Scheme 1: Preparation of succinate NCO prepolymers and cast thermoplastic polyurethanes made with BDO chain

extender showing polyester polyol (2), NCO prepolymer (3), and resultant TPU (4).

10 BioAmber is not making any recommendation or endorsements of one preparation method over the other, but feels it is necessary to highlight these differences as they could be a source of difference between values reported here and those from other sources such as [Ref. 4-6].

(3)

(2)

(4)

Page 9

Table III shows that the work life and gel times associated with conversion of the prepolymer (3) to the TPU (4) were essentially unchanged regardless of which polyol was used. Cured sheets of PU materials were compared against industry benchmark polymers based on AA-BDO and PTMEG and also are shown in Tables III and IV. All urethanes exhibited a similar Shore A hardness range of 80-85. There were some differences in property trends the researchers noted and several of these will be discussed in the following sections.

Table III: Key characterization data for NCO and CPU polymers used in this study MPD = 2- methyl 1,3 propane diol, BDO = 1,4 butane diol, HDO = 1,6 hexane diol, PDO = 1,3 propane diol

Polyol Abbre-viation in CPU

SA-MPD SA-MPD/BDO

SA-HDO/BDO

AA-HDO/BDO SA-PDO AA-PDO AA-

BDO PTMEG

Diol Type 2-methyl

1,3-propane

diol

50 mol% 2-methyl 1,3-

propane diol and

1,4 butane diol

50 mol% hexane diol

and 1,4 butane diol

50 mol% hexane diol

and 1,4 butane diol

Susterra™ bio-

based 1,3 Propane

diol

Susterra™ bio-based

1,3 Propane

diol

1,4 butane

biol

poly tetra-methylene

oxide

Prepolymer Processing Information work life (min) @ 100°C

13 14 14 13 16 20 14 11

CPU Soft Segment Tg onset (°C)

-12 -18 -37 -42 -20 -39 -41 -40

CPU Soft Segment Tg midpoint (°C)

-8 -13 -33 -39 -20 -38 -37 -34

CPU Hard Segment Tm (°C)

162/168 159/165 168 170 170 170 166 166

Bio-Based Carbon in CPU (C14 by ASTM D6866)

29 29 26 0 63 21 (estimated) 0 0

Page 10

Mechanical and chemical properties

Standard ASTM test techniques [4-6] were used to compare mechanical properties for the polyurethanes; results are summarized in Table IV and several properties are highlighted in Figures 3-7. The cast TPUs studied had excellent mechanical properties compared with benchmarks based on AA-BDO and PTMEG. Stress/strain curves from tensile testing are shown in Figure 3; measured tensile and strain values are listed in Table IV and graphed in Figure 411. Interestingly, with the SA-PEP systems the TPUs exhibited very-good tensile stress at break properties in contrast to typical trends with conventional TPUs where break stress/strain values are opposite (higher break stress at lower elongation). Researchers found that polymers based on SA-HDO/BDO and SA-PDO (Figure 4) had higher or similar tensile break values and higher elongations than benchmark AA-based systems.

Test results also showed that resistance to a number of common solvents for TPUs based on SA-PEPs exceed those of benchmark TPUs based on adipate polyester or PTMEG. Solvent absorption (in wt%) from room-temperature samples of the various urethanes are shown in Table VII. Note the very-low oil pickup for TPUs based on SA-PEPs.12

Mechanical and chemical testing (Tables IV, VI, and VII) of the bio-based TPUs generally showed that SA-HDO/BDO and SA-PDO exhibited a desirable combination of mechanical strength, abrasion and tear resistance, and resistance to common solvents.

11 The stress/strain data in Table IV are average values reported by Troy Polymer Inc. The reported standard deviations are considered accurate. The graphed stress/strain curves in Figures 4 and 9 are from a single data set selected by the author from one of the 5 repeats that was consider closest to the average values listed in Table IV. However the values in the Table IV should be considered more accurate and should be used for comparative analysis of the values between the different polyurethanes. 12 The urethanes made with SA-MPD and SA-MDP/BDO polyols exhibited higher relative solvent swell compared to others in this series. As presented in TPE Summit Conference, Vienna Austria, Dec. 4, 2014, polyurethanes made with AA-MDP and AA-MDP/BDO respectively had 7 day solvent swell values as follows: MEK: 125% vs. 78%, Ethyl acetate: 94% vs. 68%, Toluene: 62% vs. 50%, IRM 903: 2.0% vs. 2.4% respectively. These swell data again show that SA polyols tend to improve the solvent resistance compared to similar adipate based systems.

Page 11

Table IV: Select mechanical properties for NCO and CPU polymers used in this study MPD = 2- methyl 1,3 propane diol, BDO = 1,4 butane diol, HDO = 1,6 hexane diol, PDO = 1,3 propane diol

Polyol Abbreviation in CPU

SA-MPD SA-MPD/BDO

SA-HDO/BDO

AA-HDO/BDO SA-PDO AA-PDO AA-

BDO PTMEG

Diol Type 2-methyl 1,3-

propane diol

50 mol% 2-methyl 1,3-propane diol

and 1,4 butane diol

50 mol% hexane diol

and 1,4 butane diol

50 mol% hexane diol

and 1,4 butane diol

Susterra™ bio-based

1,3 Propane

diol

Susterra™ bio-based

1,3 Propane

diol

1,4 butane

biol

poly tetra-methylene

oxide

Properties at RT (before aging)

Tear strength, kN/m (ASTM D624)

41 ±1.8 72 ±2.1 89 ±2.4 91 ±1.5 92 ±1.4 95 ±2.2 127 ±4.4 90 ±1.2

Tensile Stress (ASTM D412)

Tensile Strength @ break at RT, MPa

49.5 ±4.8 47.9 ±2.5 55.4 ±1.8 53.2 ±3.2 62.2 ±5.9 55.6 ±4.4 60.5 ±7.9 44.9 ±2.6

Tensile Strength @ 50%, MPa 3.5 ±0.1 3.6 ±0.1 4.7 ±0.1 4.9 ±0.2 4.6 ±0.0 4.9 ±0.1 7.4 ±0.3 6.0 ±0.1

Tensile Strength @ 100%, MPa 4.9 ±0.1 5.0 ±0.1 6.2 ±0.1 6.4 ±0.2 6.1 ±0.0 6.4 ±0.1 8.5 ±0.2 7.9 ±0.1

Tensile Strength @ 300%, MPa 9.7 ±0.2 10.4 ±0.1 12.4 ±0.2 13.6 ±0.6 11.7 ±0.1 13.6 ±0.2 17.4

±0.4 17.7 ±0.2

Stain @ break at RT, % 1011 ±57 887 ±22 855 ±25 730 ±26 940 ±50 777 ±38 780 ±70 548 ±24

Hardness (Shore A) 85 85 85 86 87 86 89 87 Density (g/cm3) 1.21 1.24 1.2 1.18 1.25 1.18 1.18 1.09

Taber Abrasion ASTM D4060-10 (H22 Wheel, 500 g, weight loss (mg loss after 5,000 cycles) 6 10 <2 <2 <2 <2 <2 <2

Bashore Rebound (%) 10 16 38 55 27 50 50 48

Page 12

Figure 3: Representative stress/strain curves for the initial mechanical properties of the TPUs used in this study

Figure 4: Tensile and elongation at break values of TPUs (standard deviation based on the average of 5

measurements)

Page 13

Figure 5: Representative DSC data for polyurethanes made with SA-PDO and AA-PDO Measuring Tg and Tm

As noted in the previous section, DSC was used to measure Tg and Tm values for the materials in this study. Figure 5 shows a typical DSC trace for a polymer made with SA-PDO vs. that for a conventional polymer made with AA-PDO. Note that the thermal transitions detected are consistent with a phase-separated thermoplastic elastomer (TPE), and specifically that both a Tg associated with thermal motion of the polyester polyol soft block as well as the melting transition associated with urethane hard-block linkages were found. Figure 6 shows how TPUs made with SA polyester soft blocks have a higher Tg than benchmark AA-PEPs, as would be expected. On the other hand, use of a polyol with a mixed glycol (e.g. SA-BDO/HDO) made it possible to lower Tg to less than -30°C. It is generally seen that Tg decreases as the number of carbons between ester groups increase; it also is generally seen that Tg increases where pendent alkyl groups are added because the length of the chains interrupt interchain-chain alignment and reduce interchain van der Waals forces, making the chains "less sticky" than a linear system. Such characteristic behavior can be seen by comparing the Tg values for TPUs made with SA-MPD, SA-MPD/BDO, and SA-HDO/BDO. The polyester soft block's Tg decreases to -33°C from a range of -8°C to -13°C. Also notable, all the TPUs exhibited similar melting endotherms (≈165°C), which indicates that the thermal transition associated with the MDI hard block is not unduly influenced by the phase-separated polyester segment. Tm values are summarized in Table III.

Page 14

SA-PDO based TPUs

Another interesting comparison can be made when looking at the stress/strain behavior of polymers made with the branched glycol (2-methyl-1,3-propane diol (2-MPD)) vs. that for SA-PDO. Here polyols have 3 carbons between each ester linkage (vs. ≈3.5 carbons found between ester linkages with SA-BDO/2-MPD). As previously seen in Table I, all polyols in this study were low-melting solids or liquids at RT. The pendent methyl group in 2-MPD seems to decrease end polymer physical properties in much the same way that pendent methyl groups (with similar melt indices) in polyethylene and/or polypropylene do. However, this structural group also beneficially reduces low-strain tensile values (often erroneously called modulus at 50% and 100% strain) for the polymer. To achieve softer urethane coatings for synthetic leather, for example, researchers seek such lower tensile values as an initial indicator of the material's flexibility as well as softness. Table VII indicates that TPUs based on SA-MPD have the lowest tensile values at 50% and 100% strain for the series tested in this study, and furthermore these values are even lower than those for the PTMEG system. In contrast, TPUs made with SA-PDO have the next-lowest stress values at low strain, but this polymer also was found to have much improved solvent, tear, and abrasion resistance in combination with low temperature transitions vs. those produced from branched polyols. Furthermore, polymers based on SA-PDO compared well to similar polymers made with AA-PDO, thereby suggesting that the SA-PDO system could bridge performance gaps in current commercial offerings by producing polymers characterized as being softer or more flexible, while at the same time offering improved solvent resistance and containing 63% renewable carbon content.

Figure 6: Soft-block Tg for TPUs in study

Page 15

Figure 7: Percent retention of original tensile properties after exposure to humidity (7 day, 70° C, 95% RH)

Hydrolysis resistance

It would not be unreasonable to predict that TPUs produced from polyesters would be prone to hydrolytic degradation of the aliphatic ester, followed by chain scission, loss of molecular weight, and subsequent reduction in entanglements of the polymer chain. By measuring changes in tensile stress properties (before and after exposure to hydrolysis conditions), it should be possible to determine if significant hydrolysis of the ester linkages occurs in such materials. Samples from the study were exposed to 95% RH at 70°C for 7 days and changes in tensile values before and after exposure were compared and are shown in Figure 7. It also is useful to compare changes in tensile break values as well as tensile values at 100% and 300% strain, since standard deviation values are much smaller than in the break values. Further, if significant reductions in molecular weight were seen, it also would be detected at such low tensile strain values. However, as shown in Table IV and Figures 7-8, no significant difference in mechanical properties was noted.13 Researchers noted that changes in physical properties of SA-based TPUs were very similar to those seen for comparable AA-based TPUs made under similar conditions for these unstabilized polyurethane systems.

13 In a companion study presented at the UTECH-NA Sustainable Automotive Conference, June 3, 2015, Charlotte, NC, hydrolysis data were presented on similar systems tested at 65° C, 95% RH for 15 days. The mechanical property retention data for polyurethanes made with adipates and succinates were essentially the same as shown in Figures 7 and 8 and seemed more consistent with plasticization and not degradation.

Page 16

Table V: Hydrolysis resistance for NCO and CPU polymers used in this study MPD = 2- methyl 1,3 propane diol, BDO = 1,4 butane diol, HDO = 1,6 hexane diol, PDO = 1,3 propane diol

Polyol Abbre-viation in CPU SA-MPD SA-

MPD/BDO SA-

HDO/BDO AA-

HDO/BDO SA-PDO AA-PDO AA-BDO PTMEG

Diol Type 2-methyl

1,3-propane diol

50 mol% 2-methyl 1,3-

propane diol and

1,4 butane diol

50 mol% hexane diol

and 1,4 butane diol

50 mol% hexane diol

and 1,4 butane diol

Susterra™ bio-based

1,3 Propane

diol

Susterra™ bio-based

1,3 Propane

diol

1,4 butane

biol

poly tetra-

methylene oxide

Hydrolysis Resistance (7 days, 70C 95% RH)

Tensile @ 100% Strain (MPa) 4.5 ± 0.12 4.4 ±0.17 4.7 ±0.13 5.1 ±0.18 5.2 ±0.34 5.4 ±0.14 5.7

±.0.24 6.3 ±0.19

Tensile @ 300% Strain 9.1 ±0.23 9.6 ±0.05 9.5 ±0.30 10.8 ±0.33 9.7 ±0.20 11.0 ±0.30 12.8

±0.12 15.0 ±0.51

Tensile @ Break 43.9 ±4.3 46.1 ±3.5 35.0 ±1.8 37.3 ±3.0 41.6 ±1.6 35.5 ±1.3 44.4 ±1.6 34.2 ±2.8

Retention of Tensile Modulus (%) @

100% Strain 75% 78% 82% 82% 77% 80% 67% 80% 300% Strain 74% 78% 79% 78% 76% 80% 75% 84%

Break 82% 89% 78% 82% 88% 77% 90% 97%

Table VI: 7-day solvent weight absorption (%)

Polyol in PU SA-MPD

SA-MPD/ BDO

SA-HDO/ BDO

AA-HDO/BDO SA-PDO AA-

PDO AA-BDO PTMEG

Solvent Wt% Solvent Swell after 7 days

Water 1.3 1.5 1.2 1 1.6 1.3 1.1 1.7

5% NH4OH 2.3 2.7 1.2 1 2 1.4 1 1.8

Ethanol 14 13.2 12.7 14.4 9.8 13 10.8 27.4

MEK 11410 8410 73.9 75.1 59.5 68.4 73.5 69.3

Toluene 39 35 47.5 55.4 22.4 41 47.3 56

Ethyl Acetate 83 68 64.3 63.2 48.6 58.6 60.4 56.1

IRM 903 (63° C) 0.5 0.3 2.1 4.4 0.5 2.2 3.2 9.7

Page 17

Table VII: Property summary of SA-PEPs with pendent methyl groups vs. SA-PDO *Calculated bio-based carbon content based on 33% bio content in AA-PDO and at 66% polyol content

MPD = 2- methyl 1,3 propane diol, BDO = 1,4 butane diol, HDO = 1,6 hexane diol, PDO = 1,3 propane diol

PU → SA-MPD SA-MPD/BDO SA-PDO AA-PDO PTMEG

Bio-carbon content % (ASTM D6866) 29 29 63 20* 0

Tg-mid-point (°C) -8 -13 -20 -40 -34

Tensile Stress @ 50% Strain (MPa) 3.5 3.6 4.6 4.8 5.9

Tensile Stress @ 100% Strain (MPa) 4.9 5 6 6.4 7.9

Tear Resistance Die C (kN/m) 41 72 92 95 90

Taber Abrasion (mg weight loss) 6 10 < 2 < 2 < 2

Break Stress (MPa) 49.5 47.9 62.2 55.6 44.9

Break Strain (%) 1011 890 940 777 550

7-day Ethanol Swell (%) 14 13 10 13 27

7 day Toluene Swell (%) 39 35 22 41 56

7-day Oil Swell @ 63C (%) 0.5 0.3 0.5 2.2 10

Page 18

Figure 8: Representative stress-strain curves of polyurethanes made with SA-PDO and AA-PDO

Representative stress/strain curves for TPUs made from SA-PDO and AA-PDO are compared in Figure 8. Before and after stress/strain curves show a drop in tensile values at any given strain but without significant loss of elongation. Such mechanical property changes are far more similar to those seen with plasticization of a polymer network than what would be expected with molecular weight reduction. That is not to say that hydrolysis of the SA- and AA- systems is not occurring, or that such systems were exempt from the thermodynamics of esters. Rather, the researchers merely note that SA- and AA- systems prepared by the prepolymer method did exhibit similar behavior under test conditions used in the study. Also they have shown that formulation and processing conditions influence the polymer network and therefore the physical properties (including hydrolysis resistance) of the resultant polymers. Urethane coatings produced from PUDs tend to have poorer hydrolysis resistance than solvent-borne urethane coatings because the dispersing acid used to make PUD increases polymer hydrophilicity, hence increases water-polymer interactions while reducing hydrolytic resistance [3]. Eventually the hydrolytic stability of the TPUs (or any other resin) will need to be investigated under conditions that mimic those of the end-use application, since many factors (including stabilizer use, processing conditions, and even part geometry) all can impact this critical property.

Page 19

Conclusions BBSA is a chemical building block that offers the plastics industry an opportunity to both

improve performance and increase sustainability simultaneously for any urethane polymer (thermoset or thermoplastic) as well as any polyester polymer (thermoset or thermoplastic). Results from this study have shown that BBSA-based polyester polyols can be a viable alternative to petrochemical-based polyester polyols. They offer significant formulation flexibility as well as opportunities to improve solvent resistance ― in some cases without loss of other key properties (e.g. strength, tear resistance, and even hydrolysis resistance). Tradeoffs seem primarily to be higher Tg, higher rebound, and higher viscosity for SA polyols used to produce bio-based TPUs, and even then the viscosity changes remain within a useful processing range. Furthermore, selection of the glycol and acid SA/AA mixtures can modify even these properties, as the study also showed.

Recent capacity increases (to 30,000 tonnes/year) at a facility in Sarnia, Ontario, Canada near Detroit mean that commercial quantities supplied by a robust supply chain are now available as industry begins to ramp up its demand for the production of more sustainable urethane polymers. As familiarity with succinate systems increases, so too will opportunities for commercial adoption of bio-based urethanes in a variety of application areas.

References 1. Coggio, W. D.; Mullen, T.; Hevus, I.; Croes, K.; Kingsley, K.; Webster, D.C.; Zweep, N., Bio-Based

Succinic Acid: A Versatile Building Block for Performance Driven PUD and Coatings; Presented at American Coatings Conference Session #2, Biobased Coatings, Atlanta, GA April 2014.

2. Coggio, W. D.; Schrock, A.; Thompson, B.; Ulrich, K., Modified Polyester Polyol Succinates Derived from Bio-Based Succinic Acid and Branched Glycols: Influence of Glycol Structure on Tg, Tm and Process Viscosity; Presented at UTECH-NA, conference Charlotte, NC, June 4, 2014.

3. Coggio, W.D., Succinic Acid; A Bio-based Building Block for Higher Performance Polyurethanes Dispersion for Coatings. Presented at the Waterborne Coatings Conference, New Orleans, LA February 10, 2015. In this study, thin PU films based on SA and AA PEP and made with H12-MDI were cast from a PUD and cured. The mechanical properties were evaluated after exposure to 60° C, 95% humidity and for 3, 7, 19 days. Our data indicated the PU based films based on these systems showed substantial degradation after 7 days and were fully decomposed after 19 days. But again the SA and AA systems exhibited very similar behavior under these test conditions

4. Miller, R.; Janssen, R.; Theunissen, L.; Evaluating the Properties of Performance of Susterra® 1,3 Propanediol and Biosuccinium™ Sustainable Succinic Acid in TPU Applications. Council for Polyurethane Industry (CPI) Conference Atlanta, GA Sept 2012.

5. Janssen, R.; Theunissen, L.; Evaluating the Properties of Performance of Biosuccinium™ Sustainable Succinic Acid Based Copolyester Polyols in TPU Applications. Council for Polyurethane Industry (CPI) Conference Phoenix, AZ, Sept 2013.

6. Janssen, R.; Theunissen, L.; Investigation of Hydrolytic Stability of Thermoplastic Polyurethanes based on Biosuccinium™ Sustainable Succinic Acid Containing polyester polyols. Council for Polyurethane Industry (CPI) Conference Dallas, TX, Sept 2014.