Printer Friendly

Feasibility of determining the amount of oil in SB type rubber by pulsed NMR spectroscopy.

Analysis of rubber additives for quality control is routinely performed during production of styrene-butadiene rubber. The first step in the analysis is extraction of the rubber sample and gravimetric measurement of the percent total extractables, which consist of oil, soap, organic acid and antioxidant. The latter three are measured separately; the combined mass percent of oil and low-molecular-weight polymer is then calculated using mass balance. This combined mass percent is thus subject to the cumulative errors in the actual determinations. Typically, the presence of the polymer is ignored so that the extraction test method overestimates the amount of oil.

The purpose of this work was to develop a technique that allows the determination of rubber additives without solvent extraction, thus eliminating the expense of sample preparation and solvent disposal. A pulsed nuclear magnetic resonance (pulsed NMR) spectroscopy method for doing this has been evaluated. Applications for which this method is suitable are discussed.

In pulsed NMR (refs. 1-3), the sample is irradiated with a short (1-2 [micro]s) intense rf pulse at the nuclei's resonance frequency. The nuclei initially absorb the rf electromagnetic radiation and the magnetization vector is rotated by 90 [degrees]. Over time, the nuclei return to their equilibrium state, resulting in a decay of the observed nuclear magnetic field. The decaying signal, called the free induction decay (FID), is collected as a function of time and yields the NMR signal for the sample. The FID signal is observed as an exponential decay curve, figure 1.

[Figure 1 ILLUSTRATION OMITTED]

In the spin-echo technique, a second pulse (180 [degrees], or inversion pulse) is applied to the sample. The second pulse increases the amplitude of the oil signal and an echo signal is obtained. The amplitude of this echo signal is proportional to the quantity of the oil in the oil-extended polymer (figure 2). The spin-echo technique has been applied to many problems in the elastomer field, including determination of dispersion of filler in various elastomers (refs. 4-6), molecular weight and molecular weight distribution (ref. 7), glass transition temperature and diffusion of extender oil in various elastomers (ref. 8).

[Figure 2 ILLUSTRATION OMITTED]

Experimental

Instrumentation

A low-resolution, 20 MHz proton pulsed [sup.1]H-NMR spectrometer (Oxford Instruments) with 125-mm diameter magnet pole faces and 18 mm diameter sample tube was used in this study. The spectrometer contained a dual-channel phase sensitive detector with a programmable low-resolution filter. The instrument conditions used were an analyses time of 36 s (36 scans, 1 s/scan), a [T.sub.2] of 1 ms and a spin-echo measurement of 160 points (0.2 [micro]s/point). The NMR pulse was optimized using the oil-extended rubber with the highest oil content. A proton control analysis (PCA) was used to set up a Hann echo experiment for the measurement of the oil content in the rubber.

Sample preparation and analysis

Calibration standards containing various levels of naphthenic and aromatic oils were prepared by adding the appropriate amount of the emulsified oil to styrene-butadiene latex. The oil-extended rubber was coagulated using salt acid. The oil content of the standards ranged from 1 to 36%. The composition of the oils used in this study is shown in table 1.
Table 1 - composition of extender oil

Oil type Aromatic Polar Saturated Asphaltenes
 (%) com- hydrocarbons (%)
 pounds max, (%)
 (%)

Aromatic 68 15 20 0.1
Naphthenic 36 3 62 0.1


A sample (200-300 g) of each standard was milled on a 12 x 16 inch Farrel-type mill at a temperature of 120 [degrees] F prior to analysis. After cooling to room temperature, a piece of the milled rubber (1.0-3.0 g) was weighed to the nearest 1 mg. The rubber sample was placed in the NMR tube using a stopper tool. To eliminate temperature effects, the samples were pre-heated to the temperature of the NMR sample chamber (40 [degrees] C).

Results and discussion

As with many other techniques, the method parameters must be defined before meaningful data can be generated. In the case of quantifying the amount of oil in oil-extended styrene-butadiene rubber using pulsed NMR, sample mass, reproducibility, styrene content of the polymer, carbon black and temperature are all critical parameters.

Sample mass

Various samples (0.2 to 4 g) of styrene-butadiene rubber extended with 30% aromatic process oil were analyzed to determine the optimum sample size for analysis. A plot of detector response (signal/mass) vs. sample mass yielded a straight line for samples up to 2.0 g (figure 3). For larger samples, the detector response continuously declined, suggesting a non-linear response of the radio frequency detector. These results show the need to determine the optimum sample size prior to performing routine sample analysis.

[Figure 3 ILLUSTRATION OMITTED]

Reproducibility

Replicate determinations (10) of percent mass of oil in a sample (same tube loading) of styrene-butadiene rubber extended with approximately 30% aromatic process oil, using the sample mass (1.5 [+ or -] 0.001 g) gave (30.36 [+ or -] 0.17) mass percent. Replicate determinations (10) of percent mass of oil from the same bale (different tube loading) of rubber gave (30.69 [+ or -] 0.57) mass percent. Analysis of the same sample by the extraction (ASTM D-1419) method gave a value of 30.46%.

Comparison of the instrument reproducibility determined above with sample-to-sample reproducibility showed that the sample reproducibility is the major contributor to the standard error. Sample reproducibility is affected by the sample preparation techniques and the sample homogeneity, while instrument repeatability depends only on the stability of the instrument signal (signal/noise). There was a significant decrease in the magnitude of the residual error when samples were milled prior to analysis. This is not unexpected, since the oil is more uniformly distributed in the milled rubber, and samples selected for analysis are more homogenous.

Determination of oil content of oil-extended rubber

Aromatic and naphthenic oils are frequently added to styrene-butadiene robber to enhance physical properties with concentrations ranging normally from no oil to 70 phr.

To evaluate the validity of the test method for different extender-oils, a series of styrene-butadiene rubber samples extended with various levels of aromatic and naphthenic oils was analyzed by NMR. The same samples were also analyzed by the extraction test method (ASTM D-1419). The mass % oil determined by the NMR method was a linear function of the mass % determined by the ASTM method for both naphthenic oils ([r.sub.2] = 0.9866) and aromatic oils ([r.sup.2] = 0.9970) equations 1 and 2, respectively.

Naphthenic:

(1) Mass % oil (signal/mass, NMR) = 454.1 x (mass % oil, ASTM) + 12,186

Aromatic:

(2) Mass % oil (signal/mass, NMR) = 437.09 x (mass % oil, ASTM) + 11,963

Differences in the NMR signal between aromatic and naphthenic process oils .were determined using two rubber samples, one extended with 17% of aromatic oil and the other with the same amount of naphthenic oil, figure 4. The styrene content of the rubber in both cases was 23.5%. The calibration curve used to determine the percent oil by NMR was generated using rubber samples extended with aromatic oil. The percent oil obtained was then compared to that obtained by the extraction test method. Figure 4 clearly shows that calibration standards extended with the appropriate type of oil are needed. This is not unexpected, since [sup.1]HNMR spectrometers measure the total number of protons in a sample and the number of protons per unit mass of oil is different for different process oils.

[Figure 4 ILLUSTRATION OMITTED]

Effects of polymer structure in the oil determination

The technique employed in this study uses differences in proton spin-spin relaxation times of polymer and oil phases to determine the amount of oil in the oil-extended rubber. The fact that the relaxation times of the polymer protons are much smaller than those of the oil proton is the criterion used to differentiate the signals originating from the two phases.

The effect of the change in the styrene content of the polymer on the relaxation times of the oil protons was determined by comparing the spin-echo signal of styrene-butadiene rubber (23.5% styrene content) and polybutadiene rubber, each extended with aromatic oil and with naphthenic oil. Figures 5 and 6 show that the NMR signal for the oil protons is significantly affected by the changes in the styrene content of the polymer. Equations 3 and 4 are linear regression fits for samples of styrene-butadiene rubber (23.5% styrene) and polybutadiene rubber extended with various levels of aromatic oil, respectively.

SBR (aromatic oil):

(3) Signal/mass = 437.09 (mass % oil, ASTM) + 11963

EBR (aromatic oil):

(4) Signal/mass = 183.31 (mass % oil, ASTM) + 13,359

[Figure 5 & 6 ILLUSTRATION OMITTED]

With all the samples, the NMR response was a linear function of the oil content (ASTM-D1419) of the sample. With both aromatic and naphthenic oils, the NMR response (slope of the NMR signal/mass vs. % oil [ASTM]) was stronger with SBR, relative to EBR. The magnitude of the spin-echo signal for the oil was increased with increase in the styrene content of the polymer. Also, the effects of the higher styrene content of the polymer on the signal/mass ratio were more pronounced at higher oil content.

Effects of carbon black

In addition to the process oil, carbon black is also routinely added to styrene-butadiene rubber in order to enhance physical properties. Styrene-butadiene rubber primarily used in tires benefits from the addition of carbon black through greater tensile strength, as well as wear resistance.

Attempts were made to use the pulse NMR technique described earlier to determine the mass % oil in oil-extended carbon black masterbatch. Linear regression fits of carbon black masterbatch samples extended with various levels of process oil show that the standard deviation of the technique may be too high to use the technique for determining the oil content of carbon black materbatches.

Temperature effect

Results of this work show that different types of process oils and different polymers have significant effects on the calibration curve, making it necessary to use a separate calibration curve for each type of oil and type of polymer.

High temperature experiments were performed to see if differences in the polymer and oil can be minimized to the point where a single calibration curve could be used to determine mass % oil in the rubber. Experiments were performed at 40 [degrees] C and 110 [degrees] C. The elevation of temperature did not overcome the differences (signal/mass) in the responses of the different process oils (figure 7).

[Figure 7 ILLUSTRATION OMITTED]

Conclusion

Analysis of rubber additives for quality control in production laboratories is routinely performed by solvent extraction (ASTM-D1419). This technique is labor intensive and employs large amounts of organic solvents. Results from this test are not available in time to be used as an effective tool for quality control. We have investigated the possibility of replacing the current extraction method with NMR.

The bench top pulsed NMR used in this study is simple to use and results are available in a short time after the sample to be analyzed is brought to the lab. Most measurements are initiated and completed within a few minutes. Sample preparation consists of filling a standard laboratory test tube. No extraction or any other sample preparation are necessary to determine the amount of oil in the polymer. It minimizes the drawbacks of the conventional extraction method, while offering benefits: speed, safety (eliminates toxic solvents) selectivity (responds selectively to the protons of the oil molecules).

The NMR signal amplitude for every type of oil is different, which means the operator has to use a separate calibration curve for every product extended with a different type of oil. At times, different batches of the same oil may even vary sufficiently to warrant frequent monitoring of the signal amplitude of every batch of oil and alter the calibration factor accordingly. A complication of the technique is that the NMR signal for the oil is affected by the styrene content of the polymer. This may require the use of a separate calibration curve each time there is a significant change in the styrene content of the polymer. Finally, the proposed NMR technique will not get rid of the solvent extraction test completely, since, in addition to the oil, there may still be a need to measure the amount of soap, organic acid and antioxidant in the rubber.

Acknowledgements

"Feasibility of determining the amount of oil in SB type rubber by pulsed NMR spectroscopy," is based on a paper given at the April, 1999 Rubber Division meeting. "Styrenic thermoplastic elastomers," is based on a paper given at the October, 1998 Rubber Division meeting.

References

[1.] And, T.T. and Roberts, J.D. (1979). Plastics and Rubber: Materials and Applications, 4, 138.

[2.] Fukushima, E. and Rolder, S.B.W. (1981). Experimental Pulse NMR, Addision Wesley Publishing Company Inc.

[3.] Farra, T.C. and Becker, E.D. (1971). Pulse and Fourier Transform NMR, London Hydon and Son Ltd.

[4.] Svoboda, J. and Odchnal, M., Markromol. Chem., 164, 295 (1973).

[5.] Kuznetsor, B.V. and Marchenko, G.N., Vysokomol. Soedirs, Ser. A, 17, 1777 (1975).

[6.] Kontos, E.G. and Slichter, W.P., J Polym. Sci. 61, 61 (1962).

G.N. Ghebremeskel, Nathan Westendorf and Cebron Hendrix, Ameripol Synpol Corp.3
COPYRIGHT 1999 Lippincott & Peto, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1999, Gale Group. All rights reserved. Gale Group is a Thomson Corporation Company.

Article Details
Printer friendly Cite/link Email Feedback
Comment:Feasibility of determining the amount of oil in SB type rubber by pulsed NMR spectroscopy.
Author:Hendrix, Cebron
Publication:Rubber World
Geographic Code:1USA
Date:Sep 1, 1999
Words:2205
Previous Article:Styrenic thermoplastic elastomers.
Next Article:Compounding Ingredients Price List.
Topics:


Related Articles
Instrumentation concerns similar to those of a century ago.
CB's role in compound curing behavior.
Characterization of physical properties of SBR/carbon black masterbatch.
New ethylene standards for EPDM.
Controlling the iridescent bloom in EPDM body sealing applications.
Standardization of EPDM characterization tests for QC and specification purposes.
Compound analysis, formula reconstruction.
Longevity of NR structural bearings.
Powder rubber -- a new raw material generation for simplifying production - pt. 2.
Chlorination of NR: surface and failure analyses in a post-vulcanization bonded composite. (Cover Story).

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters