The Effect of Human Error on the Temperature Monitoring and Control of Freeze Drying Processes by Means of Thermocouples.

Monitoring product temperature is mandatory in a freeze-drying process, in particular in the process development stage, as final product quality may be jeopardized when its temperature trespasses a threshold value, that is a characteristic of each product being freeze-dried. To this purpose thermocouples are usually inserted in some of the vials of the batch to track product dynamics. The position of the thermocouple inside the vials strongly affects the reading of the temperature evolution during the freeze-drying process and, thus, it is necessary to place them in the right position, in such a way that correct information about product temperature is obtained. In this work, at first, the probability of the operational error resulting into a wrong positioning of the thermocouple inside the vial has been estimated experimentally. Then, the effect of this error has been assessed in terms of risk of exceeding the limit temperature in the primary drying step. Both 4R and 10R vials have been considered, and the investigation evidenced that the probability of incorrect thermocouples placement can reach 30% for 10R vials, and about 32% for 4R vials. These probability values increase, respectively, to 47 and 39% when the trays containing the vials are shifted to their final position. Then, through IR thermal imaging it has been possible to evaluate the temperature gradients in a vial, pointing out that the temperature difference between the product at the center of the vial, where the thermocouple is supposed to be, and that of the wall, that is quite often measured by the thermocouples, can be about 1°C. Therefore, associated to each thermocouple reading there is a probability distribution of product temperature. These figures can be used to assess the risk of exceeding the limit temperature in a freeze-drying process and, thus, to quantify suitable safety margins when evaluating thermocouple readings to take into account the operational errors, given a risk tolerability criteria.