ETD

Archivio digitale delle tesi discusse presso l'Università di Pisa

Tesi etd-06282019-094304


Tipo di tesi
Tesi di laurea magistrale
Autore
TORTORELLI, SONIA
Indirizzo email
snotortorelli@gmail.com
URN
etd-06282019-094304
Titolo
Experimental and modelling study of mass-transfer and fluid dynamic characteristics of Mellapak M452.Y in distillation columns
Dipartimento
INGEGNERIA CIVILE E INDUSTRIALE
Corso di studi
INGEGNERIA CHIMICA
Relatori
relatore Prof.ssa Brunazzi, Elisabetta
Parole chiave
  • distillation column
  • pressure drop
  • Mellapak 452.Y
  • HETP
  • different diameter column
Data inizio appello
22/07/2019
Consultabilità
Non consultabile
Data di rilascio
22/07/2089
Riassunto
Experimental and modelling of mass transfer and fluid dynamic characteristics of Mellapak 452.Y in distillation columns

This work focuses on the comparison of the pressure drop and the HETP values of Mellapak 452.Y obtained in the distillation columns with an inner diameter of 300 mm, R300, and with an inner diameter of 150 mm, R150. The data were measured under atmospheric pressure and total reflux conditions using methanol/n-propanol mixture. The data on R300 column were measured whitin this thesis, while, the R150 data were acquired from the workplace database.

Pressure drop and HETP were measured and elaborated in function of the F-factor which depends on the liquid and the vapour phases flows. Compared parameters (pressure drop, HETP) are linked to the behavior of the entire packed bed. The composition of phases changes all the way along the bed, consequently the F-factor changes as well. Therefore, both measured parameters are linked to the composition range between the head and the bottom of the column.

The aim of this thesis is to propose a proper F-factor able to describe the bed performance most relevantly and to suggest a method to convert the measured data according to the reference composition of distillate mixture, under certain reference conditions. Hydraulic and transport data were recalculated under the same reference conditions.

The direct transfer of the hydraulic and the transport data from the testing columns of small diameter to the large industrial columns involves a risk of inaccuracy which increases with an increasing change of the scale. The possibility of usage of the small distillation columns to collect data for direct transfer to design a large industrial column needs to be well verified by comparing the data it provides with those measured in the larger column of established institution. The packed bed performance with a structured packing is different for small and large columns. The difference in the diameter manifests itself in the different influence of the column wall on the packed bed performance. For the small diameter columns, the negative effect of the liquid phase wall flow or bypassing vapour (coming along the walls and not through the bed) is more important. For the large diameter columns, it is more difficult to achieve required uniform radial distribution of the liquid and of the vapour. For small columns, a higher proportion of the inclined channels of the structured packing ends up by the wall of the column, and a smaller portion of the channels ends at the boundary between the adjacent packing elements. For the large columns the opposite holds.

Recalculation of the R300 HETP data to the single composition (equimolar mixture) using revealed composition dependence significantly reduced data scatter and thus confirmed meaningfulness of the procedure. Similar procedure applied to the R150 HETP data, unfortunately, did not reduce the scatter and led to conclude that the source of the experimental data scatter was different. The HETP values shown a different trend with F-factor for both columns. The separation efficiency achieved in R150 was worse under lower loads, perhaps due to the undesirable fluid dynamic phenomena such as wall flow or initial liquid/vapour maldistribution over the column cross section. However, the separation efficiency achieved at higher loads was equal in both columns.
The recalculated pressure drop data to the equimolar composition were found similar for both columns. The only difference was visible for lower F-Factor values; this was probably due to the method of measurement in R150 column which did not give accurate results.
Overall, it seems that both columns provide very similar hydraulic and separation efficiency data under higher loads, conditions under which the columns tend to be industrially operated.
The results allow to draw conclusions about the transferability of hydraulic and transport data between columns R300 and R150 mutually and the possibility to use these data to design the industrial scale columns.
File