Polyurethane applications for the vibrating needle curemeter.

Polyurethane applications for the vibrating needle curemeter

The ability to monitor the curing characteristics of liquid polymer formulations, for example polyurethanes, is vital to the needs of quality control, trouble-shooting and product development. Of course, one technique which is widely used is the measurement of gel time. This is simple to perform and supplies numerical data so that comparisons between systems can be made. However, such single point data may be of little value outside routine quality control. The method may serve to indicate that two cures are different but provides little further to distinguish what the differences are. Closer monitoring of the physical properties of a curing formulation provides more information about the curing characteristics of the formulation.

Viscosity is probably the physical property most widely monitored in the study of the cure of mixes which are initially free-flowing. However, it should be recognized that a curing liquid mix develops elasticity as the molecular network builds up. Thus, a simple viscometer cannot monitor reliably beyond the early stages of the cure.

The limitation of such measurements to non-elastic systems is a serious one. Undoubtedly, the technologist would like to produce a complete cure profile, as can be accomplished for conventional solid rubber vulcanization. In the latter case, rheometers or curometers are used, but most of these appear to be wholly unsuited to handling a free-flowing liquid mix. Even when such an instrument (a Wallace-Shawbury curometer) was modified to retain a free-flowing liquid, the monitoring responses of the instrument proved inadequate for study of the early, i.e. viscous, stages of the cure (ref. 1).

Such problems were recognized at Rapra in the late 1970s and initiated the development of an entirely new instrumental technique. This instrument in its earliest guise (ref. 2) has been used extensively in Rapra's own developments on cure control, and to support other consultancy work on liquid polymer cures. This is the Rapra Vibrating Needle Curemeter (VNC), which is capable of production notionally complete cure profiles of many different types of formulations (e.g. polyurethanes, liquid polysulphides and unsaturated polyester resins).

The instrument

The Vibrating Needle Curemeter (VNC) operates by suspending a steel needle in the formulation to be monitored. This needle is vibrated vertically by a small electrodynamic vibrator driven by a single generator (see figure 1). Resistance to its movement is ultimately recorded on a chart recorder or personal computer. This allows the instrument to be unattended during cure monitoring. A trace analogous to a Monsanto rheometer cure trace for solid rubber can be obtained.

The VNC responds to changes both in the viscosity of curing formulations before gelation and subsequent changes in the stiffness of the formulation after gelation. Changes in viscosity are detected by the consequent changes in the damping of the needle's vibration. Development of stiffness or elasticity in the formulation causes a change in the resonance frequency of the vibration system. Both these effects cause a change in the VNC's output voltage.

Figure 2 demonstrates the VNC's response to silicone liquids of different viscosities. Figure 3 illustrates the change in the resonance peak (i.e. voltage vs. frequency) for the needle, during the cure of a polybutadiene polyol with IPDI at 60 [degrees] C. The figure shows how the magnitude of the peak decreases in the early stages of the cure (compare the curves for 2 and 10 minutes). This is due to the increased damping of the needle's vibration as viscosity increases. During the later stages of the cure the position of the resonance peak changes to higher frequency (compare the curves for 75 and 115 minutes). This latter effect results from the increase in the stiffness or elasticity of the formulation during cure.

Data handling

For ease of data handling, numerical data can be derived from VNC traces. Figure 4 shows how such data (T10, T80, T95 and T100) can be obtained from a cure trace obtained with the VNC operating at 40 Hz. The cure being monitored was that of Diorez 571 (100 parts) with Hyperlast isocyanate 2875/000 (30.8 parts), in the presence of dibutyltin dilaurate (0.13 parts).

The cure was repeated several times and the numerical data obtained for each cure are listed in table 1. These data demonstrate the reproducibility of the VNC, the standard deviation (or spread) of the results being below 5% of their mean.

Table : Table 1 - VNC data for a PU formulation

Cure         T10      T80      T95     T100
 no.        (sec.)   (sec.)   (sec.)   (sec.)
  1           53       81      100      199
  2           49       82      102      206
  3           51       81      105      211
  4           51       81      105      211
  5           51       81      109      220
  6           52       80      101      204
  7           49       79       99      200
  8           50       79       99      199
Average       51       81      103       -
 spread        1.3      1        3       -
% spread       3        1        3       -

Comparison with viscosity measurement

As mentioned previously, the progress of a cure can be monitored using a Brookfield Viscometer. Figure 5 compares the build-up in viscosity with time, with the cure trace obtained with the VNC operating at 40 Hz.

The cure being monitored is that of a PU formulation (Diorez) PR1, 100 parts; Hyperlast isocyanate 2875/000, 21.5 parts and tributlytin oleate, 0.2 parts). This shows that the VNC can monitor the cure well beyond the gel time, even when operating at 40 Hz. Monitoring with a Brookfield Viscometer is unreliable, if not impossible, after the onset of gelation.

By offering the facility to monitor through the gel point, the VNC brings a new dimension to liquid cure monitoring. The whole process of cure, from liquid through to solid, is presented on a single trace. That single trace can provide information on pot life, application time, cure time, etc., and a convenient format for rapid comparison between samples.

Application of the VNC to polyurethanes

By monitoring the cure of liquid polymers using the VNC, the effect of the various components in a given formulation can be easily assessed. The examples given show how the VNC can be used to monitor delay in cure, changes in cure profile and property development of polyurethane elastomers.

Effect of catalyst type

An hydroxyl-terminated polyether formulation, containing fillers and a plasticizer, has been cured with a polymeric MDI in the presence of two different catalysts. The catalysts were either an organomercury salt or an amine.

The VNC traces, shown in figure 6, show that the formulation containing the organomercury catalyst has a useful delay period of around five minutes, followed by a rapid cure.

In contrast, the amine catalyzed cure has only a short delay period (one minute), followed by a slower cure. Thus, the VNC has clearly distinguished between the character of these two cures, despite the fact that the catalyst levels had been adjusted to give both formulations the same gel time.

Effect of NCO index

The NCO index of a polyurethane formulation can have a dramatic effect on the mechanical properties of the cured product. An experiment has shown that the VNC is able to distinguish between the cures of similar formulations with different NCO indices. Figure 7 shows cure traces for the cure of a hydroxyl terminated polyester (Diorez 520) with a trifunctional polymeric MDI (Hyperlast isocyanate 2975/000), in the presence of DBTL at different NCO indices: 0.9, 1.0 and 1.1. In this experiment, the traces were recorded with the VNC operating at 150 Hz, making the instrument more responsive to changes occurring the later stages of the cure.

Effect of chain extenders

The cure of three similar polyether formulations, containing increasing amounts of 1,4-butanediol chain extender and MDI, has been monitored. This was conducted using the VCN operating at 150 Hz. The formulations are described in table 2.

The increasing hardness of these formulations (see table 2) is reflected in the cure traces shown in figure 8, the traces of harder formulations reaching a lower voltage. The cure traces also demonstrate the higher activity of the harder formulations. This higher reactivity results from the greater concentration of reactive groups in these formulations.

Table : Table 2 - polyether formulations

Code A B C

Propylan D2122

  (Difunctional PPO)              100 Parts    100 Parts    100 Parts
Dibutyltin dilaurate             0.01 Parts   0.01 Parts   0.01 Parts

Hyperlast isocyanate 2875/001

  (Difunctional polymeric MDI)   46.2 Parts   61.5 Parts   77.0 Parts
1,4-Butanediol                    9.0 Parts   13.5 Parts   18.0 Parts

Shore A hardness

(after 24 hours) 36 65 72

Foams

The ability of the VNC to monitor the cure of PU foams has been demonstrated. Figure 9 shows the VNC trace of the cure of rigid foam used in refrigerator insulation. This foam was known to have a density of 32 kg/[m.sup.3] and the following reactivity:

    Cream time:    13 seconds
    String time:   35 seconds
    End of rise:   80 seconds

The VNC data (see figure 4) obtained for this cure was: T10 - 50 seconds; T80 - 58 seconds; T95 - 68 seconds; and T100 - 140 seconds.

The VNC and RIM

An initial study has demonstrated the potential of the VNC for in-mold cure monitoring.

A fast curing isocyanurate system (Dow Chemicals Spectrim MM) was injected into a preheated mold containing a preplaced glass mat, using a Cannon RIM machine. The mold had horizontal vent holes, into one of which the needle of a VNC was inserted. In this way the VNC was able to monitor a cure within the mold. A typical VNC trace is shown in figure 10.

The ability to monitor the cure of such formulations, within a molding tool, might ultimately allow the VNC to control these cure processes. For example:

* the temperature could be optimized to achieve a given extent of cure as indicated by the VNC, within a required cycle time; or

* pressure, vacuum or mold opening may be activated upon attainment of a preset extent of cure, as indicated by the VNC.

This type of control has already been demonstrated using a system which monitors the dielectric properties of a curing formulation during a molding process (ref. 3).

This type of monitoring could also be a powerful QC tool. Moldings showing defective cure may be discarded before having recourse to expensive or time-consuming testing. Also, the ability to store a characteristic cure trace of each molding made (for example by using a personal computer) would be a useful QC aid.

Conclusion

The following features of the VNC, which is an aid to process and product development, have been demonstrated:

* Continuous monitoring of the cure of liquid systems;

* Ability to detect changes in curing formulation after the gel point;

* Ability to monitor the effect of small changes in the formulation of polyurethanes, (e.g. catalyst type, isocyanate index and chain extender content);

* Possibility of in-mold cure monitoring, allowing process control and also enhancing the VNC's use in quality control;

* Software developments allowing data logging by a personal computer, enabling automatic comparison of cures and rejection of poorly cured parts without having recourse to expensive testing facilities.

Apart from the large variety of curing liquid systems to which the VNC can be applied, it is also versatile with respect to sample size and the form of the during formulation. For example, the VNC can be taken to the sample rather than vice versa, production-scale operations can be monitored, as can samples in film-form or composite.

References

1. Uri, N., Br. Poly. J., 3, 1971, 138. 2. Willoughby, B.G., Rubber World, 187, 3, 1982, 26. 3. Day, D.R., Bromberg, M.K., 30th Annual Polyurethane Technical Meeting/Conference, October 15-17, 1986.

Acknowledgements

"Chemical degradation of polyurethane" is reprinted with permission of the Society of Plastics Industry, Polyurethane Division. "Polyurethane applications for the vibrating needle curemeter" was previously presented at the Polyurethane Manufacturers Association meeting September, 1989. "Advances in NR science and technology" was previously presented at the Rubber Division meeting in Detroit, MI, October, 1989.

PHOTO : Figure 1 - the vibrating needle curemeter

PHOTO : Figure 2 - response of the VNC to silicone fluids of different viscosities

PHOTO : Figure 3 - change in resonancy frequency during the cure of polyurethane

PHOTO : Figure 4 - calculation of T10, T80 and T95

PHOTO : Figure 5 - VNC trace and increase in viscosity of a curing polyurethane

PHOTO : Figure 6 - VNC trace for a PU cured catalyzed by two different catalysts

PHOTO : Figure 7 - VNC traces of PU formulations with different iscyanate indices

PHOTO : Figure 8 - VNC traces of PU formulations of different hardness

PHOTO : Figure 9 - VNC trace of the cure of a PU foam

PHOTO : Figure 10 - VNC trace of the cure of Spectrim at 50 [degrees]C; monitored in the RIM mold