Structural Analysis of the Terminal Groups in Commercial Hevea

Subscriber access provided by Washington University | Libraries

Article

Structural analysis of the terminal groups in commercial Hevea natural rubber by 2D-NMR with DOSY filters and/ or multiple-WET methods using ultrahigh-field NMR Muneki Oouchi, Jinta Ukawa, Yoshitaka Ishii, and Hideaki Maeda Biomacromolecules, Just Accepted Manuscript • DOI: 10.1021/acs.biomac.8b01771 • Publication Date (Web): 12 Feb 2019 Downloaded from http://pubs.acs.org on February 13, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

Structural analysis of the terminal groups in commercial Hevea natural rubber by 2D-NMR with DOSY filters and/or multiple-WET methods using ultrahigh-field NMR Muneki Oouchi,†* Jinta Ukawa,‡ Yoshitaka Ishii,†§ Hideaki Maeda†♯

†NMR Science and Development Division, RIKEN SPring-8 Center (RSC), and NMR Facility, CLST, RIKEN, 1-7-22 Suehiro, Tsurumi, Yokohama, Kanagawa 230-0045, Japan ‡Toyo Tire Corporation, 3-10-1 Yato, Kawanishi, Hyogo 666-0131, Japan §School of Life Science and Technology, Tokyo Institute of Technology, 4259 Nagatsuta, Yokohama 226-8503, Japan ♯JST-Mirai Program, Japan Science and Technology Agency, 1-7-22 Suehiro, Tsurumi, Yokohama, Kanagawa 230-0045, Japan.

ACS Paragon Plus Environment

1

Biomacromolecules 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 36

ABSTRACT. The terminal groups of natural rubber (NR) are widely believed to play a crucial role in defining the excellent mechanical and other physical properties of processed NR products. Despite their presumed importance, the chemical structures of the terminal groups are elusive in widely used NR species with a high degree of polymerization, such as Hevea natural rubber (H-NR). In previous studies, structural analysis by solution NMR has been carried out on the terminal units of NR after chemical treatment involving chemical alterations, such as deproteinization with enzymes and other chemicals. However, there is concern that such chemical treatments may alter the properties of the terminal units. In this study, we established an NMR-based approach to analyze the structures of the terminal units in commercial H-NR without any chemical treatments, or with only a mild treatment of some samples, such as acetone extraction for removing the impurities. To suppress the signals of low-molecular-weight impurities, we have developed methods combining DOSY-based diffusion filters with multiple-WET (MWET) 2D-NMR, which we introduced previously to suppress strong signals from main-


2

Page 3 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

chain of polymer and solvents [ K. Tanaka et al. Macromolecules, 2016, 49, 5750-5754]. Using the new method and MWET 2D-NMR methods with high-field NMR at a 1H frequency of 900 MHz, we observed NMR signals of the terminal units of chemically untreated commercial H-NR for the first time. The NMR results for 8 commercial H-NR samples consistently demonstrated the presence of at least 5 kinds of terminating-end (α-terminus) units of the H-NR polymer chain in addition to NMR signals for the initiatingend (ω-terminus) units. Our NMR analyses revealed for the first time that none of the αterminal groups form a phosphate ester.

INTRODUCTION

The analysis of polymer microstructures, such as terminal groups, substructures and branches, is of considerable interest in the polymer industry because the microstructures influence the intriguing properties, such as the topology, tensile strength and viscoelastic behavior, of polymers. In particular, the structural characterization of the terminal groups has been of prime importance because of their vital role in polymerization and in determining physiochemical properties.


3


Page 4 of 36

With their excellent mechanical properties, natural rubbers (NRs) are indispensable raw materials for industrial rubber products, such as tires for vehicles and aircrafts.1 Among the various raw materials of commercial polymers, high-molecular-weight NRs are one of the most important materials for which no detailed characteristics of their terminal structures are known.2-7 Nearly all commercial NRs are Hevea natural rubber (H-NR) obtained after processing the sap of Hevea brasiliensis (Para rubber tree),1-7 as shown in Figure 1a. Interestingly, even more than 80 years after the first introduction of synthetic rubber (SR),1 H-NR is generally superior to SR in terms of the mechanical and some chemical properties.1 H-NR is predominantly composed of 1,4-polyisoprene with a cis content of 100%, except at the terminal groups (Figure 1b);1,3-7 an average molecular weight of H-NR by gel permeation chromatography shows typically higher than 100 kg mol-1. To produce high-performance SRs that mimic NR, a molecular-level understanding of NR is needed. From recent studies of synthetic polyisoprene rubber, a high cis-1,4-polyisoprene content of ~99% or higher is known to improve the mechanical properties.8 In addition to stereoregularity, the structures of terminal groups in the rubber chain are hypothesized to be critical to the excellent mechanical properties of


4

Page 5 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

NR.1 In addition, more recent efforts to decode the rubber-tree genome have renewed interest in the microstructure of NR as it relates to the biochemical synthesis of NR.9

NMR is one of most powerful means to examine structures of NR and other polymers. Nevertheless, determination of the terminal structures in high-molecular-weight NR, such as H-NR, is extremely difficult because of the trace amount of the termini leading to insignificant intensity when performed even at high-field NMR at 1H frequencies (νH) of 400–750 MHz with a conventional room-temperature probe.3-5,7 The most recent NMR study has used high-field NMR at νH of 600 MHz equipped with a cryogenic probe; they have suggested the presence of 4 types of α-terminal units in H-NR latex.10 However, for those previous studies, instead of measuring commercial H-NR itself, they performed chemical treatments such as the deproteinization on NR latex with protease and SDS or saponification, and further purified the resultant samples by removing minor impurities by acetone extractions.3-5,710 Critically, the structures of terminal units may be altered by such chemical treatments, which involve chemical modifications. Therefore, it


5


Page 6 of 36

is desirable to use commercial H-NRs without additional chemical treatments to elucidate the structures of terminal units by NMR.

In our work, we introduced a new approach to identify the terminal groups of H-NR by solution NMR without any chemical treatments such as deproteinization and saponification. The H-NR samples were either dissolved directly in an NMR solvent without any treatments or the samples were dissolved in an NMR solvent after simple acetone extractions, which remove only dissolved impurities without dissolving H-NR. The major difficulty of these NMR measurements is that the low 1H-NMR signal intensities of the terminal groups are 104 times smaller than those of the main-chain. To overcome the difficulty, we used a high-field NMR system at a 1H frequency of 900 MHz equipped with a sensitive cryogenic probe. Such combinations can yield higher sensitivity and higher resolution in 1H spectra for the low-population terminal species. A remaining critical problem is that strong 1H signals from the NR main chain also reduce the sensitivity to observe the weak terminal signals because the receiver gain of the instrument cannot be matched to the level of the weak terminal signal. In our previous


6

Page 7 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

paper,11 we successfully observed the weak signals of terminal groups and substructures of synthetic polymers such as commercial poly(ethylene terephthalate) using 2D-NMR by suppressing strong 1H signals with a novel MWET(n) (multiple-WET12 with n suppressed points) method for observing substructures and end-groups.

However, we realized that in the spectra of commercial H-NRs obtained using the MWET method, the signals of low-molecular-weight (LMW) species severely overlap with those of the terminal groups, hindering the terminal analysis by this approach alone. These impurities were remained not only in the unremoved samples, raw H-NR samples, but also even in the samples removed with simple acetone extraction. Here, we propose a new approach for identifying weak terminal signals of H-NR at an unprecedented low level without chemical treatments by suppressing LMW species with the "diffusion-ordered NMR spectroscopy (DOSY)" method,13-17 while suppressing strong 1H main-chain signals with MWET(n) for 1D- and 2D-NMR. This approach is likely applicable to a wide range of NRs and other synthetic polymers.

EXPERIMENTAL SECTION


7


Page 8 of 36

Sample Preparation. The H-NRs used were commercial Ribbed Smoked Sheet No. 3 (RSS3, see Table S2 for more detail) and high-ammonia natural rubber (HANR) latex. The eight samples in Table S2 which were unhomogenized to avoid alteration were taken from four different production lots (A-D) of RSS3 in order to investigate the differences in the amount of terminal groups between samples. For example, the number average molecular weight (Mn) of production lot B was 183000 (approximately 2700 cis-1,4-isoprene units; here, Mn / Mu; Mu is the molecular weight (68) of the cis1,4-isoprene unit (C5H8)), and the weight average molecular weight (Mw) was 1560000, as determined by gel permeation chromatography (GPC) (Tosoh Corporation, Japan) and shown in Table S2. The four samples in Table S2 were purified by simple acetone extraction using an ASE-350 accelerated solvent extractor (Japan Dionex Corp. Co., Ltd.) (acetone solvent, 10.3 MPa, 10 minutes, 3 or 8 extractions). Then, the acetone-extracted H-NRs were dried by decompression to remove the acetone. The Mn of production lot B after extraction was 270,000 (approximately 4000 cis-1,4-isoprene units), and the Mw was 1600000. Samples (10-15 mg) of H-NR were separately


8

Page 9 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

dissolved in CDCl3 in 5 mm OD NMR tubes to yield concentrations of 2-3% (w/v). These samples were used to record all NMR spectra.

NMR Measurements. All 1H-NMR measurements were carried out on a 900 MHz Bruker Avance III HD spectrometer with a 5 mm cryogenic probe (TCI) equipped with a z-gradient coil with a maximum nominal gradient strength of 53.5 G cm–1 at room temperature. The sensitivity of a cryogenic probe included in the NMR system (with cold pre- or head-amplifier) is generally 3-6 times better than that of a non-cryogenic probe, and the sensitivity at 900 MHz is generally ca. 3.4 times better than that at 400MHz (although the actual enhancement depends on quality-factor of probe and other factors). The resultant high sensitivity was critical in the analysis of the terminal species of the HNR samples. 13C-NMR measurements were carried out either on a 600 MHz Bruker Avance III HD spectrometer with a 5 mm cryogenic probe (DCH) or on 900 MHz instrument with a 5 mm cryogenic probe at room temperature. All NMR data were obtained using Topspin v 3.2 or 3.5 software. With the existing wide-band 13C decoupling using chirp adiabatic pulses, decoupling side-bands remained, as shown in


9


Page 10 of 36

Figure S9a. 13C decoupling during acquisition was improved to suppress decoupling side-bands using chirp adiabatic pulses, as shown in Figure S9b.

In pulse sequences in Figure 2, MWET(n) pulses were optimized by selecting 3-5 frequencies to excite resonances from δH 7.26 (chloroform) and δH 5.12, 2.05, 2.03 and 1.68 (four frequencies of cis-polyisoprene signals).11 Sinc shape excitation pulses ((sin (x))/x, 0 < x < π)) with a 30 ms duration were used for MWET(n) 1D and 2D experiments with similar experimental conditions as in previous reports.11 The pulse programs of MWET 1D and 2D were improved for Topspin v3.5 software. Additional details of the characteristics of MWET, such as the excitation profiles and selectivity, are shown in Figure S8.11

RESULTS AND DISCUSSION

Figure 2 shows the pulse sequences with (a) MWET11,12 filter and (b) MWET and DOSY13,14 (MWET-DOSY) filter blocks. As shown in the outline of Figure 2b, in MWETDOSY-filter 1D and 2D-NMR, DOSY blocks reported as “longitudinal eddy current delay with bipolar pulse pairs (LEDBP)”13,14 were added to MWET-1D and MWET-2D such as


10

Page 11 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

MWET-1D, MWET-DQFCOSY, MWET-TOCSY, MWET-HSQC, MWET-HMBC, etc. in Figure S1. In (b), after the MWET and MWET-DOSY blocks, which respectively suppress strong isoprene signals and LMW-species signals, the signals are detected in either of 1D- and 2D-NMR, the latter of which includes DQFCOSY, TOCSY, HSQC, HMBC, etc. (see Figure S1). The MWET NMR sequence in (a) provides higher sensitivity but more pronounced LMW signals without DOSY-filter. With these sequences, we studied the terminal units of commercial H-NR samples.

Figure 3 shows (a) 1H-NMR, (b) MWET 1H-NMR, and (c) MWET-DOSY 1H-NMR spectra of commercial H-NR (3%) in CDCl3 (for more details, see Figures S2 and S3). From the MWET-DOSY 1H-NMR spectrum in (c), seven different signals are observed with assignments for 5 different terminating-end units (α-terminus; α1–α5). In the MWET 1H-NMR

spectrum in Figure 3b, the signals for α1-α5 of the terminal units in the red

circle were retained, but the analysis was not straightforward because of additional signals at δH 4.6 and 5.4 and in the high-field region at ~1 ppm, which were attributed to LMW species. In (c), these signals due to LMW species were effectively removed by the


11


Page 12 of 36

DOSY filter. The high-field 1H NMR spectrum with additional instrumental artifacts in (a) is similar to the spectrum in (b), but the sensitivity of signals, for example, at δH 4.17 (shown as blue arrow) in (a) was three-fold lower than that in (b) because of the unmatched receiver gain as previously reported.11

Although the sensitivity of signals at δH 4.17 in (c) was comparable to that of 1H NMR without MWET in (a), the MWET-DOSY sequence offers a clear advantage in that terminal-signal data can be analyzed without complications due to LMW impurities.

Figure 4 shows the overlaid MWET-HSQC (red) and MWET-HMBC (black) spectra of commercial H-NR. We identified at least five terminating-end (α-terminal) units from the spectra. Below, we briefly discuss the assignments for the α1–α5 groups and highlight the novel findings from the analysis. The probable structures of the α1–α5 groups based on our analyses are listed in Figure 1c. An analysis of the initiating-end (ω-terminal) unit is also shown in Figure 1b and described below.

α-terminal α1. In the MWET-HSQC spectrum (red) in Figure 4, a 2D 13C/1H correlation of the α-terminal α1-a is observed at δC 62.7/δH 4.17 (F1/F2) in Figure 4. Furthermore,


12

Page 13 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

in the HMBC spectrum (black) in Figure 4, long-range C/H correlations of the α-terminal α1-a are observed at δC 35.0/δH 4.17(α1-b), δC 145.3/δH 4.17(α1-c) and δC 173.8/δH 4.17 (α1-f), as shown in the red rectangle in Figure 4.

Previous studies presumed the presence of phosphate esters such as CH2OP5,7 or carboxylate ester10 in the α-terminus α1 of H-NR. In the case of phosphate esters such as CH2OP, the long-range two-bond coupling constant 2JCP is reported as several Hz,18 but the carbon peak of α1-a at δC 62.7 is a singlet in the 13C-NMR spectrum in Figure 4 (see expanded spectra in Figure S4-2). Moreover, in the MWET-HMBC spectrum, a long-range C/H correlation of the α-terminal α1-a correlates with the carbonyl carbon α1-f at δC 173.8/δH 4.17. As a result, it is concluded that α-terminal α1 is not a phosphate ester5,7 but a carboxylate ester10 (-CH2OCO-).

α-terminal α2. A recent study on saponified H-NR10 reported that a small fraction of the α-terminus exists as an α2 group with hydroxyl groups. We observed the corresponding signals at δH 3.71 (α2-g) without saponification treatment (Figure 3b). In the MWETDQFCOSY and MWET-TOCSY spectra in Figure S5 (see the SI), the chemical shifts of


13


Page 14 of 36

α-terminal α2-g and α2-h are identical to those of saponified H-NR, as reported. We concluded that the identification of α2 in ref. 10 is correct. However, the relative intensity of α2 varied greatly depending on the sample over a range of approximately 0-0.1 of that of α1 (see Figure S3 and Table S3).

α-terminal α3. Previous studies involving the deproteinization of NR suggested that the α-terminal α3 of H-NR is a phosphate ester.5,7,10 We observed the corresponding signals at δH 3.84 and 3.95 (α3-l) without deproteinization treatment (Figure 3). In the MWET-HSQC spectrum (red) in Figure 4, one-bond C/H correlations of the α-terminal α3-l at δC 69.2/(δH 3.95 and 3.84) are observed. Furthermore, in the MWET-HMBC spectrum (black) in Figure 4, long-range C/H correlations of the α-terminal α3-l are observed at δC 16.9 (α3-o), 32.9 (α3-m), and 174.0 (α3-p).

We further attempted to correlate the aliphatic 13C signals of α3-o, α3-m, and α3-l to the methyl 1H signal of α3-o. However, the high-field regions of the MWET-HSQC (red) and MWET-HMBC (black) spectra in Figure 5a contain numerous cross-peaks, which make the analysis very difficult due to LMW impurities. In contrast, the MWET-DOSY-


14

Page 15 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

HSQC (red) and MWET-DOSY-HMBC (black) spectra in Figure 5b clearly demonstrate C/H correlations of the α-terminal α3-o at δC 16.9/δH 0.92, and in the HMBC spectrum in Figure 5b, long-range C/H correlations of the α-terminal α3-o are evident at δC 32.9/δH 0.92 (α3-m) and δC 69.2/δH 0.92 (α3-l).

The data clearly indicate that many of the peaks in Figure 5a are attributed to LMW species, which is not obvious from the data in Figure 5a alone. Additionally, the highfield region of the MWET-DOSY-DQFCOSY and MWET-DOSY-TOCSY spectra in Figure S5 (see SI) confirmed these assignments. In the H/H correlation spectra of MWET(5)- DOSY-COSY and MWET(5)- DOSY-TOCSY, the peaks at δH 0.92 (α3-o) clearly correlate at δH 1.77 (α3-m), and 3.84 and 3.95 (α3-1).

Notably, the carbon peak of α terminal α3-l at δC 69.2 is a singlet in the 13C-NMR spectrum in Figure 4 (see expanded spectra in Figure S4-2). Therefore, it is confirmed that α3 cannot be a phosphate ester. The α-terminal α3-1 correlates with the carbonyl carbon α3-p at δC 174.0/(δH 3.95, 3.84) in the HMBC spectrum in Figure 4. Contrary to


15


Page 16 of 36

that reported previously, 5,7,10 it has been confirmed that the α-terminal unit α3 links to carboxylate esters.

α-terminal α4. A small amount of α-terminal α4, which is not a carboxylate ester as at α3 but a hydroxyl group, was recently reported after saponification treatment.10 We also observed signals at δH 3.42 and 3.51 (α4-q) in the spectra of samples that did not undergo saponification treatment shown in Figure 3. It is concluded that α-terminal α4 exists.

α-terminal α5. In addition to the aforementioned 4 kinds of α-terminal units, α-terminal α5 signals of H-NR were observed at δH 4.56 in Figure 3. In Figure 4, a C/H correlation of the α-terminal α5-u is observed at δC 60.9/δH 4.56, and long-range C/H correlations of the α-terminal α5-u are located at δC 119.4 (α5-v), 142.3 (α5-w) and 173.8 (α5-z). This α-terminal α5 of H-NR is similar to the α-terminal unit identified in the relatively lowmolecular-weight NR from the mushroom Lactarius volemus.19 This is the first time that this type of α-terminal unit has been reported for H-NR, which has been studied for decades.


16

Page 17 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

α-terminal α6.1a,3,4,7 Unfortunately, clear signals corresponding to α-terminal α6 were not observed in the commercial H-NR analyzed. However, the very weak hydrogen signals at δH 4.09 in Figure 4 (see Figure S7) may be signals from α-terminal α6-u’, which are 0.47 ppm upper-field from α5-u (δH 4.56).

Overall, in commercial H-NR, we observed at least 5 units, α-terminal α1–α5, as shown in Figure 1c, whose structures were analyzed. Quantification of these units was straightforward. The relative amounts of the α1, α3, α4, α5-terminals determined by 1H integration were approximately 45%, 100%, 25%, and 15%, respectively, as summarized in Table S3, where the amount was normalized to α3. Notably, the relative amount of α2 was generally low (< 7%) but varied from sample to sample. This novel approach to quantifying different terminal groups may be useful for characterizing the degradation or aging of the NR chain in rubber products.

ω-terminal units.1a,4,7 In the 1H spectrum shown Figure 3c (see expanded spectra in Figure S3), three characteristic hydrogen signals at δH 1.58(s), 1.60(s) and 1.61(s) were observed. In the MWET-DOSY-HSQC spectrum in Figure S6 (see SI), a C/H correlation


17


Page 18 of 36

of those three peaks is observed at δC 15.9/δH 1.58 (ω-6’’), δC 16.0/δH 1.60 (ω-1”) and δC 16.0/δH 1.61 (ω-6”), respectively. In the HMBC spectrum in Figure S6, one of the methyl groups, which was correlated at δC 16.0/δH 1.60 (ω-1”) in MWET-DOSY -HSQC, was also correlated at δC 39.8, 124.4 and 131.3. Additional, in the MWET-DOSY HMBC spectrum in Figure S6, long-range C/H correlations of the other two methyl units correlated at δC 16.0/δH 1.61 (ω-6”) and δC 15.9/δH 1.58 (ω-6”) are observed at δC 39.7, 124 and 135.0. Our analysis of these signals suggests that the ω end groups are characterized by a single dimethylallyl terminal unit with two trans-isoprene units linked to cis-1,4-polyisoprene, as identified for the ω-terminus of relatively low-molecularweight NR obtained from L. volemus, for example.6,19

CONCLUSIONS

Using novel MWET-DOSY 2D-NMR combined with our previously developed MWET 2D-NMR method and high-field NMR, we convincingly demonstrated that structural analysis of the terminal groups of H-NR and other rubbers with a high degree of


18

Page 19 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

polymerization is possible without any chemical treatments. Unlike previous studies, we observed at least 5 α-terminal units of α-terminal α1-α5, including newly identified α5 group. In addition to the newly identified α5 group, we reported the lack of a linkage of the α3 group to a phosphate group. It was confirmed that the ω-terminal groups link to the dimethylallyl terminal, which links to two trans-isoprene units that link to the cispolyisoprene units. As our samples are commercial H-NR that is rubber raw material of tire, we expect broad applications of the NMR method developed here. Such detailed structures of the terminal groups of H-NR will enable the artificial or biochemical synthesis of H-NR for the production of H-NR at the industrial scale. The developed approach is likely applicable to a wide range of other NRs and other synthetic polymers.


19


CH3 CH3

C

C

CH3 CH2

CH2

C

H

CH3

C

CH2

H

CH2

C

CH2

n

2

-terminal group

α1, α2 α3, α4 α5, etc

H

C

Page 20 of 36

-terminal

Commercial Hevea natural rubber sheet Para rubber tree (Hevea brasiliensis)

Tire, Engine mount, etc.

Abstract

a

Para rubber tree Commercial Hevea natural (Hevea brasiliensis) rubber (RSS3 sheet)

+acid ・・・・・sheet ・・ Sap・・smoked ・・ 1’’

Part of the sheet

1’

4’ 3’ CH2 5’

C

C

H

α1 e d

CH3

7’ CH2 CH26’ C 8’

CH3 2’ C

H

CH2

α2

O

CH2

s

t

r

CH

α5 CH2

CH3

q

OH

C

3

CH2 2

2

α1, α2 α3, α4 α5, etc

H

C

1

4

CH2

n

k j

C b CH2 C CH2c CH2a O f R

α4

α-terminal

5

6’’CH3

CH3

c

+SRs, C, S, etc. ・・・

ω-terminal group

b

Tire, Engine mount, etc.

α3

CH2

C h CH2 CH2 CH2g OH i

y CH3 v C

CH2

x

w

C

H

u

CH2 O

z

n

α6

O C

CH2

R

o

m CH

O CH2

l

O

C

p

R

CH3

CH3 C CH2

H C

u’

CH2 OH

Figure 1. (a) Production scheme of commercial H-NR sheet and H-NR rubber products. (b) Chemical structure of H-NR with the ω-terminal structure suggested in this study. (c)


20

Page 21 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

Structures of α-terminals (α1-α6) suggested in this study. The 1H/13C chemical shifts of the α-terminals are provided in Table S1 (see the Supporting Information (SI)).

a

Multiple-WET(n)

1D-NMR

IR, T2 filter etc.

WET

1H

2D-NMR

13C

13C

DQFCOSY, TOCSY HSQC, HMBC, etc.

dec

PFG

b

DOSY-filter

Multiple-WET(n) WET

13C

PFG

13C

τ1

τ2

τ1

IR, T2 filter etc.

2D-NMR

LEDBP τ1

1H

1D-NMR

LEBBP, STEBP, etc.

τ1

τ3

DQFCOSY, TOCSY HSQC, HMBC, etc.

dec

g1 g1

g1 g1

Figure 2. Outline of the MWET 1D/2D (a) and the MWET and DOSY filtered (MWETDOSY) 1D/2D (b) pulse sequences. These sequences are shown in more detail in the SI in Figure S1. For the DOSY filter, the LEDBP sequence was adopted. "g1, τ1-τ3" are important parameters for implementing the DOSY-filter method.


21


Page 22 of 36

α1-a α1-e α3-l α2-g α5-u α4-q

c

#1

ω6" ω1"

#2

b

5

a

＊

3

4,1

＊

＊

Figure 3. 1H-NMR spectra (CDCl3 at 900 MHz) of the 3% H-NR solution of sample A after simple acetone extraction. (a) 1H-NMR with the receiver gain (RG) of 6.35. “*” indicates artifact signals from the instrument. (b) MWET(5) 1H-NMR with RG of 203. (c) MWET(5)-DOSY 1H-NMR with RG of 203. τ1=1.7 ms, τ2=40 ms, τ3=5 ms, g1=ca. 42.8 G cm-1 at 1.5 ms (see Figure 2). The five suppressed points are marked with green arrows in (b). #1 was -CH2CH2OCO-, and #2 was -CH2CH2OH, but these points were not clearly assigned.


22

Page 23 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

1H

α1-e α2-k

α5-u

α6?

α1-a

α3-l

α2-g

α4-q

13C

o b

b

t

m

j d u

k e

a

g l

q

v w

c

z

f

p

Figure 4. Overlaid 2D 13C/1H correlation spectra of a 3% H-NR solution of sample A (without acetone extraction) in CDCl3 at 900 MHz obtained with MWET-HSQC (red) and MWET-HMBC (black) using the same suppressions described in Figure 3b. 13C spectra at 151 MHz (1H 600 MHz) in Figure S4-1 .


23


b

a 1H

13C

Page 24 of 36

α3-o, α4-t

1H

o, t

o, t m, r ｎ l q

Figure 5. High-field 1H region of the 2D 13C/1H correlation spectra of sample D (without acetone extraction) in CDCl3 (a) without and (b) with the DOSY filter at 900 MHz. (a) Overlaid MWET-HSQC (red) and MWET-HMBC (black) spectra. (b) Overlaid MWETDOSY-HSQC (red) and MWET-DOSY-HMBC (black) spectra. “τ1-τ3 and g1” are mentioned in Figure 3.

ASSOCIATED CONTENT

Supporting Information.


24

Page 25 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

The Supporting Information is available free of charge on the ACS Publications website.

The chemical shift table, experimental section, pulse sequences, and additional NMR data (PDF), and Bruker pulse programs.

AUTHOR INFORMATION

Corresponding Author *(M.O.) Email: [email protected]

Notes The authors declare no competing financial interest.

ACKNOWLEDGMENT All NMR experiments were performed at the RIKEN NMR Facility as an activity of the “NMR Open Sharing and Platform Program” supported by the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan. The work was also


25


Page 26 of 36

supported, in part, by the JST-Mirai Program (grant No. JPMJMI17A2, Japan) to Y. I. The authors would like to thank Mrs. Kimiko Tanaka and Miss Mami Maeda of Mitsui Chemical Analysis & Consulting Service Inc. for use of the MWET characteristics in the Supporting Information.

REFERENCES (1) (a) van Beilen J.B.; Poirier Y. Establishment of new crops for the production of natural rubber. Trends in Biotechnology 2007, 25, 522–529. (b) Cornish k. Similarities and differences in rubber biochemistry among plant species. Phytochemistry 2001, 57, 1123–1134. (c) Rodgers B.; Wadeel W.H.; Klingensmith W. in Encyclopedia of Polymer Science and Technology, 4th ed., Vol.12 (Ed.H.F.Mark) John Wiley & Sons, Hoboken, NJ 2014.

(2) (a) Burfield D.R. Epoxy groups responsible for crosslinking in natural rubber.

Nature 1974, 29-30. (b) Moir G. F.J. Ultracentrifugation and staining of Hevea latex. Nature 1959, 1626-1627.


26

Page 27 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

(3) (a) Tanaka Y.; Aik-Hwee E.; Ohya N.; Nishiyama N.; Tangpakdee J.; Kawahara S; Wittsuwannakul R. Initiation of rubber biosynthesis in Hevea brasiliensis: Characterization of initiating species by structural analysis. Phytochemistsy 1996, 41, 1501-1505. (b) Tangpakdee J.; Tanaka Y. Long-chain polyprenols and rubber in young leaves. Phytochemistsy 1998, 48, 447-450.

(4) (a) Eng A.H.; Kawahara S.; Tanaka Y. Trans-isoprene units in natural rubber. Rubber Chem. Technol., 1994, 67 159 -168. (b) Tanaka Y. Structural characterization of natural polyisoprenes: Solve the mystery of natural rubber based on structural study.

Rubber Chem. Technol. 2001, 74, 355-375. (c) Mekkriengkrai D.; Sakdapipanich J.T.; Tanaka Y. Structural characterization of terminal groups in natural rubber: Origin of nitrogenous groups. Rubber Chem. Technol. 2006, 79, 366-380.

(5) Tarachiwin L.; Sakdapipanich J.; Ute K.; Kitayama, T.; Bamba T.; Fukusaki E.; Kobayashi A.; Tanaka Y. Structural characterization of α-terminal group of natural rubber. 1. Decomposition of branch-points by lipase and phosphatase treatments.

Biomacromolecules 2005, 6, 1851-1857.


27


Page 28 of 36

(6) Tanaka Y.; Mori M.; Ute K.; Hatada K. Structure and biosynthesis mechanism of rubber from Fungi. Rubber Chem. Technol. 1990, 63, 1-7.

(7) (a) Tanaka Y.; Tarachiwin L. Recent advances in structural characterization of natural rubber. Rubber Chem. Technol. 2009, 82, 283-314. (b) Thuong N.T.; Yamamoto Y.; Nghia P.T.; Kawahara S. Analysis of damage in commercial natural rubber through NMR spectroscopy. Polym. Degrad. and Stab. 2016, 123, 155-116. (c) Saito T.; Klinklai W.; Kawahara S. Characterization of epoxidized natural rubber by 2D NMR spectroscopy. Polymer 2007, 48, 750-757.

(8) Kaita S.; Doi Y.; Kaneko K.; Akira C.; Horiuchi A. C.; Wakatsuki Y. An efficient gadolinium metallocene-based catalyst for the synthesis of isoprene rubber with perfect 1,4-cis microstructure and marked reactivity difference between lanthanide metallocenes toward dienes as probed by butadiene-isoprene copolymerization catalysis. Macromolecules 2004, 37, 5860-5862

(9) (a) Epping J.; Deenen N.; Eva Niephaus E; Stolze A; Fricke J; Huber C; Eisenreich W; Richard M. Twyman R.M; Dirk Prüfer D.; Gronover C.S. A rubber transferase


28

Page 29 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

activator is necessary for natural rubber biosynthesis in dandelion. Nature Plants 2015, 1, 15048 (b) Yamashita S.; Yamaguchi H.; Waki T.; Aoki y.; Mizuno M.; Yanbe F.; Ishii T.; Funaki A.; Tozawa Y.; Miyagi-Inoue Y.; Fushihara K.; Nakayama T.; Takahashi S. Identification and reconstitution of the rubber biosynthetic machinery on rubber particles from Hevea brasiliensis. eLife 2016, 5, 19022. https://elifesciences.org/articles/19022. (c) Takahashi S.; Lee H.; Yamashita S.; Koyama T. Characterization of cisprenyltransferases from the rubber producing plant Hevea brasiliensis heterologously expressed in yeast and plant cells. Plant Biotechn. 2012, 29, 411-417.

(10) Kitaura T.; Kobayashi M.; Tarachiwin L.; Kum-ourm H.; Matsuura A.; Fushihara K.; Ute K. Characterization of natural rubber end groups using high-sensitivity NMR. Macromol. Chem. Phys. 2018, 219, 1700331.

(11) (a) Tanaka K.; Oouchi M.; Hayashi F.; Maeda H.; Waki H. Structural analysis of the end groups and substructures of commercial poly(ethylene terephthalate) by multiple-WET 1H/13C NMR. Macromolecules 2016, 49, 5750-5754. (b) Oouchi M.; Tanaka K.; Waki H.; Hayashi F.; Maeda H. Structural analysis of the end groups and


29


Page 30 of 36

substructures of commercial poly(ethylene terephthalate) by multiple-WET 1H NMR. 58th Experimental Nuclear Magnetic Resonance Conference (2017) Asilomar CA.USA.

(12) (a) Ogg, R. J.; Kingsley, R. B.; Taylor, J. S. WET, a T1- and B1-insensitive watersuppression method for in vivo localized 1H NMR spectroscopy. J. Magn. Reson., Ser. B 1994, 104, 1−10. (b) Smallcombe, S. H.; Patt, S. L.; Keifer, P. A. WET solvent suppression and its applications to LC NMR and high-resolution NMR Spectroscopy. J.

Magn. Reson., Ser. A 1995, 117, 295−303.

(13) (a) Johnson C.S. Diffusion ordered nuclear magnetic resonance spectroscopy: principles and applications. Progress in NMR Spectroscopy 1999, 34, 203–256. (b) Morris K. F.; Johnson C.S. Diffusion-ordered two-dimensional nuclear magnetic resonance spectroscopy. J. Am. Chem. Soc. 1991, 114, 3139-3141.

(14) (a) Wu D.; Chen A.; Johnson C.S. An improved diffusion-ordered spectroscopy experiment incorporating bipolar-gradient pulse. J. Magn. Reson., Ser. A 1995, 115, 260264. (b) Cotts R.M.; Hoch J.R.; Sun T.; Markert T. Pulsed field gradient stimulated echo


30

Page 31 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

methods for improved NMR diffusion measurements in heterogeneous system. J. Magn.

Reson. 1989, 83, 252-266.

(15) (a) Jayawickrama D. A.; Larive C. K.; McCord E. F.; Roe D. C. Polymer additives mixture analysis using pulsed-field gradient NMR spectroscopy. Magn. Reson. Chem. 1998, 36, 755-760. (b) Ohuchi M.; Meadows P.; Horiuchi H.; Sakai Y.; Furihata K. Dynamics of sodium and lithium counter-ions and water molecules in cation-exchange resins as shown by NMR spectroscopy. Polymer J. 2000, 32, 760-770.

(16) (a) Wu D.; Chen A.; Johnson C.S. Three-dimensional diffusion-ordered NMR spectroscopy: The homonuclear COSY–DOSY experiment. J. Magn. Reson., Ser. A 1996, 121, 88-91. (b) Jerschow A.; Müller N. 3D diffusion-ordered TOCSY for slowly diffusing molecules. J. Magn. Reson., Ser. A 1996, 123, 222-225. (c) Barjat H.; Morris G.A.; Swanson A. G. A three-dimensional DOSY–HMQC experiment for the highresolution analysis of complex mixtures. J. Magn. Reson. 1998, 131, 131–138. (d) Buevich A. V.; Baum J. Residue-specific real-time NMR diffusion experiments define the association states of proteins during folding. J. Am. Chem. Soc. 2002, 124, 7156-7162.


31


Page 32 of 36

(17) (a) King A. W. T.; Mäkelä V.; Kedzior S. A.; Laaksonen T.; Partl G. J.; Heikkinen S.; Koskela H.; Heikkinen H. A.; Holding A. J.; Cranston E. D.; Kilpeläinen I. LiquidState NMR Analysis of Nanocelluloses. Biomacromolecules 2018, 19, 2708−2720. (b) Nilsson, M. The DOSY Toolbox: A new tool for processing PFG NMR diffusion data. J.

Magn. Reson. 2009, 200, 296−302. (c) Crutchfield, C. A.; Harris, D. J. Molecular mass estimation by PFG NMR spectroscopy. J. Magn. Reson. 2007, 185, 179−182.

(18) (a) Gray G. A. 13C nuclear magnetic resonance of organophosphorus compounds. I. Diethyl phosphonates. J. Am. Chem. SOC. 1971, 93, 2132-2140. (b) Davisson V. J.; Woodside A.B; Neal T. R.; Stremer K. E.; Muehlbacher M.; Poulyer D. Phosphorylation of isoprenoid alcohols. J. Org. Chem. 1986, 51, 4768-4779.

(19) Thoung N.T.; Kosugi K.; Kawahara S. Structural characterization of rubber from

Lactarius Volemus through 2D-NMR spectroscopy. Kautsch. Gummi Kunstst. 2015, 68, 26-32.


32

Page 33 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

1D-NMR

One dimensional NMR

2D-NMR

Two dimensional NMR

DEPT135

Distortionless enhancement by polarization transfer.

COSY

Correlation spectroscopy

DQFCOSY　

Double quantum filtered COSY

TOCSY

Total correlation spectroscopy

HSQC

Heteronuclear single-quantum correlation spectroscopy

HSQC-TOCSY

HSQC with TOCSY

HMBC

Heteronuclear multiplebond correlation spectroscopy

edited HSQC

Multiplicity-edited HSQC


33


INADEQUATE

Page 34 of 36

Incredible natural-abundance double-quantum transfer

experiment

WET

Water suppression enhanced through T1 effects

MWET(n)

Multiple-WET with n suppressed points

DOSY

Diffusion-ordered NMR spectroscopy

LEDBP

Longitudinal eddy current delay with bipolar pulse pairs

MWET-1D

MWET One dimensional NMR

MWET-2D

MWET Two dimensional NMR

MWET-DQFCOSY

MWET filtered DQFCOSY

MWET-TOCSY

MWET filtered TOCSY

MWET-HSQC

MWET filtered HSQC

MWET-HMBC

MWET filtered HMBC


34

Page 35 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

MWET-HSQC-TOCSY

MWET filtered HSQC-TOCSY

MWET-DOSY

MWET and DOSY filter

MWET-DOSY-DQFCOSY

MWET and DOSY filtered DQFCOSY

MWET-DOSY-TOCSY

MWET and DOSY filtered TOCSY

MWET-DOSY-HSQC

MWET and DOSY filtered HSQC

MWET-DOSY-HMBC

MWET and DOSY filtered HMBC

MWET-DOSY-HSQC-TOCSY

MWET and DOSY filtered HSQC-TOCSY


35


1D-NMR 2D-NMR

Multiple-WET Multiple-WET CH3 CH3

C

C

CH2

H

C

1D-NMR 2D-NMR

DOSY-filter

CH3 CH2

C H

CH3 CH2

-terminal group

Page 36 of 36

CH2

2

C

α1, α2 α3, α4 α5, etc

H C

CH2

n

-terminal

Commercial Hevea natural rubber sheet Para rubber tree (Hevea brasiliensis)

Tire, engine mount, etc.

TOC


36

Structural Analysis of the Terminal Groups in Commercial Hevea

Recommend Documents