Moisture content determination of different polymers by Karl Fischer titration.

The effect of moisture on the morphology and mechanical properties of different polymers has been extensively studied by a number of researchers (refs. 1-10). The moisture content of the compounding ingredients should be well within the specification, as higher moisture content may adversely affect the performance of the polymeric goods (refs. 11-14). In some cases, a minimum amount of moisture content is required to achieve a certain property through specific chemical reaction (refs. 15-17). Therefore, it is essential to check the moisture content of the compounding ingredients, along with the other quality control parameters. There are different techniques for checking the moisture content of the chemicals. Karl Fischer (KF) titration is one of them and is widely used for chemicals (ref. 18). However, for the raw polymers, the common KF titration technique is not suitable. The reason behind this is the insolubility of the polymers in the KF solution. For raw polymers, instead of moisture content measurement, volatile loss on heating is measured (refs. 19 and 20). Another technique is the azeotropic reflux. However, both of these techniques have limitations. Volatile loss measurement gives the total amount of volatiles, including the moisture. Azoetropic reflux is suitable when the moisture content is greater than 1%. In the case of polymers, except a few, the moisture content is well below 1%. Under this condition, it is difficult to measure the exact amount of moisture.

Some advanced techniques for moisture content determination of polymeric material have been reported. David, et al (ref. 21) has reported the moisture content determination of polycarbonate and urethane. They measured moisture content by means of heating the sample under vacuum and collected in a cold trap. The vapor pressure of the water was then measured and correlated with the amount present in the sample. Camacho et al (ref. 22) measured the moisture content of nylon 6,6 by near-infrared spectroscopy and chemometrics. The determination of moisture transitions in cellulosic materials using differential scanning calorimetry has been reported by R.A. Nelson (ref. 23). Menon et al (ref. 24) has reported NMR determination of the physiological water distribution in wood during drying.

The present article describes a method for determination of moisture content of raw polymer. The technique involves heating of the polymer in an oven, carrying the total amount of volatile matter (including moisture) into a reaction vessel by dry nitrogen, titrating the moisture coulometfically by KF titration technique, leaving the other volatile matter un-reacted.

Instrument

The instrument was purchased from Kyoto Electronics Manufacturing. It consists of two modules, including a Karl Fischer moisture titrator MKC 501 (with diaphragm) and a moisture evaporator ADP 511. The flow diagram of the instrument is given in figure 1. The volumetric flow rate of the carrier (nitrogen) gas is precisely controlled by a flow controller and flow sensor. The sample is kept in the sample boat. The sample boat can be moved inside or outside of the heating zone of the heating chamber with the help of a magnetic motor system. The carrier gas (nitrogen) is passed through silica gel and a zeolite desiccant tube for drying. The dry nitrogen gas enters into the sample heating chamber and carries the total volatile matter either into the titration cell (through a heating tube) or into the gas exhaust (in purge mode). Care must be taken to ensure that the heating tube end is dipped into the anolyte solution. The cartier gas is passed through the reaction cell as long as the sample is being heated in the heating chamber. Thereafter, the gas is changed to purge mode. The pictorial presentation of the reaction cell is given in figure 2. The KF titrator has the reported accuracy of [+ or -] 1 ppm and below 0.3% refractive standard deviation (RSD) at measurement of 1 mg [H.sub.2]O in a water-methanol mixture. The moisture evaporator has [+ or -]2[degrees]C precision in temperature control.

[FIGURES 1-2 OMITTED]

Principle of measurement

In the Karl Fischer moisture content measurement, water reacts with iodine and sulfur dioxide in the presence of base and alcohol.

[H.sub.2]O + S[O.sub.2] + C[H.sub.3]OH + 3 RN + [I.sub.2] [right arrow] [RHN] S[O.sub.4]C[H.sub.3] + 2[RNH] I (1)

In the volumetric titration, iodine is added as a titrant.

In the coulometric technique, iodine is electrolytically generated in the anolyte, which contains iodide ions.

2I- [right arrow] [I.sub.2] + 2e- (2)

As long as water is present in the titration cell, the generated iodine reacts according to reaction (1). As soon as all the water reacts, an excess of iodine appears in the anolyte. The iodine is detected by the platinum electrode and the iodine production is stopped. According to Faraday's law, the quantity of iodine produced is proportional to the current generated. In equation (1), [I.sub.2] and [H.sub.2]O react with each other in proportion 1:1.

Therefore, a mole of water (18 g) is equivalent to 2 x 96,500 coulombs (or 10.72 coulombs/1 mg [H.sub.2]O). The total amount of moisture can thus be determined by measuring the total consumption of electricity.

Experimental

Chemicals used

Natural rubber (NR), styrene-butadiene rubber (SBR), polyvinyl alcohol (PVA), ethylene-propylene rubber (EPDM) and Nylon-6 were all from standard commercial sources. The Karl Fischer reagent for moisture determination was from Merck. A water standard (0.1%), traceable to SRM 2890 from NIST, was from Merck.

Analysis

Before starting the analysis, the temperature and flow rates were controlled by a digital controller. Dry nitrogen gas was passed through the heating chamber and the reaction cell to eliminate the inside moisture. Next, a pre-titration was carried out to nullify the anolyte moisture. Calculation of the moisture content was carried out according to the following equation 3:

Moisture (ppm) = (Data-drift * T - blank)/W *F (3)

Where: Data--measured water content ([micro]g); T = measuring (titration) time (sec.); W = weight of the sample (g); F = factor; Blank = blank value ([micro]g); and drift = drift before measurement ([micro]g/sec.).

Factor was determined by measuring the concentration of the standard water solution injected directly to the anolyte. The measured value was divided by the reported value (from certificate) to calculate the factor.

Blank was determined by running carrier gas (same period of time of measurement of sample) and measuring the moisture content.

Drift value was automatically calculated by the instrument when the instrument was stabilized after pre-titration of any other measurement.

The operating condition of the instrument during measurement is given in table 1.

During measurement, the polymers were cut into small pieces and taken into the sample boat. About 2-3 g of sample were taken for analysis.

The sample temperature was kept to 125[degrees]C and the sample heating time was 30 minutes. A higher heating temperature should be avoided to reduce the sample decomposition probability.

The moisture content of different polymers is given in table 2. The results were repeatable. The volatile matter of the polymers (NR, SBR, BR) as per ASTM D1278 and ASTM D5668 is also reported in table 2. The volatile matter was higher compared to the moisture content determined by Karl Fischer titration for NR, BR and SBR. Volatile matter was found to be comparable with moisture content for PVA and Nylon 6. To check the reliability of the method, a resorcinol sample was tested. The resorcinol sample was directly taken into the reaction cell and the moisture content was measured by direct titration. Under this condition, nitrogen gas was set to purge mode. The moisture content of the same sample was also determined by the heating method. The volatile matter of the same sample was measured according to ASTM D 4571 (2002). Resorcinol was taken for this study due to its solubility in the anolyte solution. The values are reported in table 3. The measured moisture content by direct titration and by heating method showed good agreement, whereas the volatile matter was relatively high.

Conclusion

This method can be applied for the moisture determination of raw polymers. However, care must be taken to avoid the sample over-heating. The polymer samples must be cut into small pieces for easy diffusion of the moisture. The method is also applicable for recycled materials like crumb rubber and polymer reclaims. The authors are using the same method for moisture determination of fillers like carbon black, clay, silica, etc. It can also be used for rubber chemical (compounding) analysis, but they are hard to dissolve in the KF solution.

References

(1.) N. Sombatsompop and K. Chaochanchaikul, Polymer International Volume 53, Issue 9, September, 2004, pp. 1,210-1,218.

(2.) S.S. Hosseini, S. Taheri, A. Zadhoush and A. Mehrabani-Zeinabad, Journal of Applied Polymer Science, Volume 103, Issue 4, February, 2007, pp. 2,304-2,309.

(3.) Y. Zhang, S.Y. Zhang and Y.H. Chui, Journal of Applied Polymer Science, Volume 102, Issue 3, November, 2006, pp. 2,668-2,676.

(4.) I. Hamerton, H. Herman, K.T. Rees, A. Chaplin and S.J. Shaw, Polymer International, Volume 50, Issue 4, April, 2001, pp. 475-483.

(5.) Z. Zhang, I.J. Britt and M.A. Tung, Journal of Polymer Science, Part B: Polymer Physics, Volume 37, Issue 7, April 1999, pp. 691-699.

(6.) E. Laredo and M.C. Hernandez, Journal of Polymer Science, Part B: Polymer Physics, Volume 35, Issue 17, December, 1997, pp. 2,879-2,888.

(7.) E. Wellisch, L. Hagan, L. Marker and O.J. Sweeting, Journal of Applied Polymer Science, Volume 85, Issue 5, August, 2002, pp. 1,031-1,039.

(8.) P.A. Dell and W.G. Kohlman, Journal of Applied Polymer Science, Volume 52, Issue 3, April 1994, pp. 353-363.

(9.) I. Arvanitoyannis, I. Kolokuris, J.M.V. Blanshard and C. Robinson, Journal of Applied Polymer Science, Volume 48, Issue 6, May, 1993, pp. 987-998.

(10.) S.H. Zeronian, M.S. Ellison and K. Alger, Journal of Applied Polymer Science, Volume 25, Issue 7, July, 1980, pp. 1,311-1,322.

(11.) N.E. Marcovich, M.M. Reboredo and M.L Aranguren, Journal of Applied Polymer Science, Volume 68, Issue 13, June, 1998, pp. 2,069-2,076.

(12.) K.L. Harrison and M.J. Jenkins, Polymer International Volume 9, September, 2004, pp. 1,985-1,998.

(13.) C. Carfagna, A. Apicella and L. Nicolais, Journal of Applied Polymer Science, Volume 27, Issue 1, January, 1982, pp.105-112.

(14.) Y. Iyenger, Journal of Applied Polymer Science, Volume 15, Issue 2, February, 1971, pp. 267-276.

(15.) A. Hunsche, U. Gorl, A. Muller, M. Knaack and Th. Gobel, Kautschuk Gummi Kanststoffe, 50, 12, December, 1997, pp. 881-889.

(16.) J. Gassan and A.K. Bledzki, Polymer Composites, Volume 18, Issue 2, April 1997, pp. 179-184.

(17.) A.D. Woolfson, D.F. McCafferty, C.R. McCallion, E.T. McAdams and M.C. Anderson, Journal of Applied Polymer Science, Volume 58, Issue 8, November 1995, pp. 1,291-1,296.

(18.) Elementary Practical Organic Chemistry, Part 3, Arthur I. Vogel Chapter XIV, 1st edition, 1987, CBS Publishers and Distributors, Delhi, India.

(19.) ASTM D1278, 2002.

(20.) ASTM D5668, 1999.

(21.) D.J. David, G.F. Baumann and S. Steingiser, Polymer Engineering & Science, Volume 2, Issue 3, July, 1962, pp. 231-236.

(22.) W. Camacho, A. Valles-Lluch, A. Ribes-Greus and S. Karlsson, Journal of Applied Polymer Science, Volume 87, Issue 13, March, 2003, pp. 2,165-2,170.

(23.) R.A. Nelson, Journal of Applied Polymer Science, Volume 21, Issue 3, March, 1977, pp. 645-654.

(24.) R.S. Menon, A.L. MacKay, J.R.T. Hailey, M. Bloom, A.E. Burgess and J.S. Swanson, Journal of Applied Polymer Science, Volume 33, Issue 4, January, 1987, pp. 1,141-1,155.

by S. Chakraborty, M. Debnath, S. Bandyopadhyay and R. Mukhopadhyay, Hari Shankar Singhania Elastomer and Tyre Research Institute (HASETRI) and A.S. Deuri, R and D Centre, J.K. Tyre ([e-mail])

Table 1--operating condition of the Instrument

Carrier gas (nitrogen) flow rate                     100 ml/min.
Heating temperature of the sample                  125[degrees]C
Heating time of the sample                               30 min.
Blank                                               0.0 [micro]g
Factor (F)                                                  0.96
Sample weight                                             2-3 g.
Measuring (titration) time          Varies from sample to sample

Table 2--moisture content of different polymers
measured by Karl Fischer titration

Polymer Type                     Measured    Average   Volatile
                         moisture content       (%)     matter
                                      (%)                  (%)

Styrene butadiene     0.110, 0.089, 0.111,      0.10       0.14
  rubber (SBR 1502)                  0.096
Natural rubber        0.392, 0.388, 0.384,      0.38       0.51
  (NR)                               0.351
Poly vinyl alcohol    2.981, 3.100, 3,062,      3.04       3.07
  (PVA)                              3.000
Poly-butadiene        0.101, 0.110, 0.087,      0.10       0.17
  rubber (BR)                        0.113
  Nylon-6             1.672, 1.701, 1.703,      1.70       1.72
                                     1.713

Table 3--moisture content determination of resorcinol

Method                         Measured value (%)     Average
                                                     value (%)

Direct KF titration    0.101, 0.101, 0.098, 0.097,       0.100
                                             0.106
Heating method         0.106, 0.100, 0.097, 0.099,       0.101
                                             0.103
Volatile loss as per   0.161, 0.153, 0.159, 0.163,       0.160
ASTM method                                  0.164