Study On The Design Of HNBR Molecular Structure And The Swelling Behavior

The excellent heat and oxygen stability and oil resistance of the hydrogenated butadiene – acrylonitrile rubber(HNBR) have made its application areas increasingly expanded. However, decrease of the ACN content does not improve the low temperature flexibility of HNBR as the ethylene unit in the chain structure becomes crystallization. One of the effective way for improving the low temperature property is to produce new ternary HNBR by introducing a proper termonomer into HNBR molecular structure to break the regular array and restrain the crystallization behavior of ethylene unit. On the basis of the solubility parameter theory, Flory-Huggins interaction parameter and FloryRehner relationship this project established a standard procedure and two prediction models, which would be used to design the molecular structure of HNBR and make new ter-HNBR with excellent high or low temperature properties.The standard procedure in this work was built up in the following way that the vapor pressure of low molecular liquids at different temperatures were measured and used to correlate the enthalpy of vaporization, which was further employed to calculate the solubility parameter according to its definition. This standard procedure was validated by acetone, cyclohexane and toluene, the results of which showed that the standard procedure was proved to be high precision and good repeatability and that the solubility parameters of low molecular liquids including solvents and monomers were comparable. The standard procedure was used to determine the solubility parameters of solvents, monomers, alkyl acrylates and alkyl propionates, and the results showed that different solvents had different solubility parameters due to their various molecular structure differences. The δ-parameters of 2-methoxyethyl acrylate(MEA) and 2-(diethylamino) ethyl methacrylate(DEAEM) are 20.89 MPa1/2 and 18.10 MPa1/2, respectively. Comparing the effect of alkyl chain length and C=C on the δ-parameter indicates that the increase of alkyl chain length decreases the δ-parameter of alkyl acrylates and alkyl propionates, and that the contribution of C=C to δ-parameter is estimated to be 0.2 – 0.3 MPa1/2.Butadiene – acrylonitrile rubber(NBR) with increasing ACN content from 18 wt% to 44 wt% was cured with the same sulfur curing system and the result indicated the similar crosslink density(S’max = 10.5 ± 0.5 d Nm) being reached for all the NBR samples. The determination of δ-parameters of NBR with the equilibrium swelling measurement indicates that the maximum swelling region corresponds to the range of δ-parameters from 18 MPa1/2 to 25 MPa1/2, and that the δ-parameter of NBR increases with ACN content showing a quatitative description between the δ-parameter and ACN molar content as δNBR = 17.4 + 10.3 [ACN]. Additionally, the regular HNBR, low temperature HNBR and new ter-HNBR were crosslinked by peroxide curing system and the results showed that both the ACN content and termonomer had influences on the crosslinking reactivities of different HNBR. Through the formulation optimization the similar crosslink density(S’max = 12.9 ± 0.5 d Nm) have been achieved for the regular HNBR, low temperature HNBR and new ter-HNBR. Then, the δ-parameters of all the HNBR were determined by equilibrium swelling measurement, the results of which indicated the maximum swelling region corresponding to the range of δ-parameters from 19 MPa1/2 to 23 MPa1/2, and the δ-parameter of HNBR increasing with ACN content showing a quantitative description between the δ-parameter and ACN molar content as δHNBR =16.8 + 11.0 [ACN].Based on the results the prediction model regarding the δ-parameter for ter-HNBR was built up as δterpolymer = δPE + α[ACN] + β[m]. As a consequence, there is an equation for low temperature as δterpolymer = 17.0 + 11.0*[ACN] + β[X], for MEA grade as δterpolymer = 17.0 + 11.0*[ACN] + 3.69*[MEA] and for DEAEM grade as δterpolymer = 17.0 + 11.0*[ACN] + 0.90*[DEAEM]. The δ-parameter determined with the prediction model fits well with that from experiment, indicating the δ-parameter of the potential HNBR with any structure and composition can be predicted by this model. Besides, the prediction model was used to investigate the effect of termonomer on the δ-parameter of ter-HNBR and the results showed that this prediction model provided a new way for selecting the optimum type, polarity and content of termonomer and as well a theoretical basis for the design of new ter-HNBR chain structure.In addition, the other prediction model with respect to the swelling property of terHNBR was established on the basis of three-dimensional solubility parameter simulated by computer program, Flory-Rehner relationship and Flory-Huggins interaction parameter. The results show that this prediction model can be used to quantitatively calculate the swelling ratio(q) of ter-HNBR. With increase of energy difference(Ra) between ter-HNBR and solvent the q decreases showing a reverse S-shape corresponding to that from experimental way. Combining the two prediction model, respectively for δ-parameter and swelling property, one can design the molecular structure of new ter-HNBR having both excellent oil resistance and low temperature flexibility which provide valuable references for the development of new ter-HNBR in the future.At the end of this work, the swelling behavior of HNBR(ACN 39wt%) in different binary solvent mixtures were investigated by using of equilibrium swelling measurement, and the Ra values were calculated correspondingly. In nonpolar-nonpolar binary solvent mixtures the swelling ratio(qm) of HNBR shows a good linear relationship with its composition, while in nonpolar-polar and polar-polar binary solvent mixtures showing non-linear relationship. This non-linearity is related to the cosolvent effect(Δq) solvent mixture influenced by the type and composition of each component, showing a maximum or minimum values for Δq. The qm of HNBR in all the solvent mixtures can be expressed qualitatively by the same qm ~ Ra curve.