43

A publication of

CHEMICAL ENGINEERING TRANSACTIONS

VOL. 69, 2018 The Italian Association
of Chemical Engineering
Online at [Link]/cet
Guest Editors: Elisabetta Brunazzi, Eva Sorensen
Copyright © 2018, AIDIC Servizi S.r.l.
ISBN 978-88-95608-66-2; ISSN 2283-9216 DOI: 10.3303/CET1869008

Extended Performance Comparison of Different Pressure

Drop, Hold-Up and Flooding Point Correlations for Packed
Columns
Verena Wolf-Zoellner*a, Frank Seibertb, Markus Lehnera
a
Montanuniversitaet Leoben,Leoben, Austria
b
University of Texas at Austin, USA
[Link]-zoellner@[Link]

Hydraulic characteristics such as pressure drop, fractional liquid hold-up and flooding point are of significant
design importance in describing random and/or structured packing performance. For this reason, accurate
modelling of these parameters over a wide operating and physical property range has to be ensured. In the
past, various research groups developed many correlations. The significant combined databases of the
Separation Research Program (SRP) at the University of Texas and the Montanuniversitaet Leoben (MUL)
provide the perfect basis to evaluate available correlations. The databases include pressure drop, hold-up and
flooding point data for 50 different random (2nd, 3rd and 4th generation) and structured packings (sheet and
grid structured packings) for different test systems under absorption and distillation conditions.
Eight correlations (Billet and Schultes 1991 and 1999, Wolf 2014, Engel 1999, SRP 1997, Delft 1997,
Verschoof 1999, Maćkowiak 2003, Stichlmair 1989) were evaluated in an initial analysis. The best predictions
were achieved with the model of Wolf 2014 and a modified model of Maćkowiak 2003 by Wolf-Zoellner.
The objective of this work is to use the large database and extend the performance investigation to further
well-known correlations available in literature (Spiegel and Maier 1992, Brunazzi 2002, Bozzano 2007, NNA-
Model of Piché 2001, GPDC of Kister 2007, Robbins 1991, Jammula 2014). Based on this evaluation, the
performance of the various correlations is compared and the application fields of the various models are
revealed. In particular, the work shows the operating ranges as well as packing types for which the models
provide reliable results but also the weaknesses of the various correlations.

1. Introduction
The reliable prediction of hydraulic properties (pressure drop, flooding point and hold-up) is essential for the
accurate design of packed columns. The pressure loss is not only important for the accurate design of the
column but also for the blower sizing. Furthermore in distillation, the pressure loss affects the pressure at the
bottom of the column and thus the temperature in the column sump which is especially important if thermally
sensitive mixtures should be separated. The flooding point defines the maximum load of counter current
operated columns. An accurate prediction of the flooding point is essential for the column diameter
determination and to prevent an over or under design. The fractional liquid hold-up prediction allows the
determination of the liquid mean residence time in the packing and is in particular important when distilling
temperature sensitive mixtures or in reactive absorption processes.
In the last few decades, many hydraulic models have been published by different research groups. The
appropriate selection of the models can be very challenging. Some authors (Kister 1992, Engel 1999,
Grabbert and Bonitz 1998, Tsai 2010) have compared a few different models but many of the available
models were not included. The aim of this work is to use the large hydraulic database obtained at the
experimental absorption plants of the Chair of Process Engineering at the Montanuniversitaet Leoben (MUL)
and of the Separation Research Program (SRP) at the University of Texas (SRP) as well as of the SRP
distillation column to examine the already published hydraulic models. Wolf-Zoellner et al. 2018 evaluated
eight correlations (Billet und Schultes 1991, 1999; Wolf 2014, Engel 1999, Stichlmair et al. 1989, Maćkowiak

Please cite this article as: Wolf-Zoellner V., Seibert F., Lehner M., 2018, Extended performance comparison of different pressure drop, hold-
up and flooding point correlations for packed columns, Chemical Engineering Transactions, 69, 43-48 DOI: 10.3303/CET1869008
44

2003, Olujić 1997 (=Delft-Model 1997), Gualito et al. 1997 (=SRP-Model 1997)). In this paper, the
performance comparison is extended to the following models: Spiegel and Meier 1992, Suess and Spiegel
1992, Brunazzi et al. 2002, Bozzano et al. 2007, Kister et al. 2007, Robbins 1991, Jammula 2014, Piché et al.
2001a and 2001b.

2. Experimental systems
Both institutions, SRP and MUL, use similar measurement columns with an inner diameter of about 420 mm to
perform the packing tests for absorption. The detailed description of the experimental set-up at MUL can be
found in Wolf et al. 2015 and of SRP in Song 2017. Comparative results obtained as well as the examined
operating range has been addressed in Wolf-Zoellner et al. 2018. An experimental SRP distillation dataset
with the cyclohexane/n-heptane (C6/C7) system was added for this evaluation. The experiments were
conducted using a 428 mm column at pressures of 0.165 bar, 0.333 bar, 1.655 bar and 4.137 bar and a
detailed description of the set-up and the operating conditions can be found in Olujić et al. 2000.

3. Correlations for hydraulic modelling

The models of Robbins and Kister were developed for random and structured packings. The NNA-model of
Piché et al. was solely developed for random packings and the models of Brunazzi, Bozzano, Spiegel and
Meier, Suess and Spiegel and Jammula are only applied to structured packings. The appropriate model
applications are shown in Table 1.

Table 1: Application fields of the examined correlations for hydraulic modelling

Random Structured Pressure Hold- Flooding
Model
packing packing loss up point
Brunazzi - yes yes yes wallis-plot
Bozzano - yes yes yes yes
GPDC - Kister yes yes yes - yes
Robbins yes yes yes - -
Spiegel and Meier - yes yes - wallis-plot
Suess and Spiegel - yes - yes -
Jammula - yes - yes yes
NNA-Model of Piché yes - yes yes -
Based on the assumptions of a falling film Brunazzi and Paglianti 1997 developed a mechanistic pressure
drop model, which includes the contributions of frictional, gravitational and acceleration losses. The model was
validated for five types of structured packings for which four empirical constants are given in Brunazzi and
Paglianti 1997. Two empirical constants can be derived from the dry pressure drop results and the two others
are used to fit the irrigated pressure drop. Unfortunately, the detailed description in determining these
empirical constants is missing in the paper. The model requires the preloading liquid hold-up as input variable.
Therefore, a second model is necessary to determine the liquid hold-up. In the present evaluation of the
Brunazzi-model, the preloading hold-up was calculated using the modified model of Maćkowiak which
performed most reliable in Wolf-Zoellner et al. 2018.
Bozzano et al. 2007 developed a completely theoretical model for the calculation of pressure drop, hold-up,
loading and flooding point for structured packings that does not need any packing empirical constants.
Unfortunately, documentation of the model is not described sufficiently enough as the required variables are
not defined. Therefore, calculation based on the given equations is not possible.
The generalized pressure drop correlation (GPDC) charts have been developed by Sherwood and Eckert and
continuously improved by different researchers over the years. The latest chart for random packings was
developed by Strigle 1994 and the latest version for structured packings is given in Kister et al. 2007. As it is
impractical to evaluate such a large database by reading each pressure drop value off a diagram, the
graphical version was converted into an equation. Tsai 2010 found an existing equation in Aspen Plus
(software developed by Aspen Tech). Its generic form is shown in Equation (1). As the Aspen software only
included constants for an older version of the chart, Tsai updated the constants in a way that the equation is
able to fit the chart given in Kister et al. 2007. To calculate the pressure drop for random packings as well the
constants for the random packing chart were derived in this work and are given together with the constants for
structured packings in Table 2. (CP =capacity parameter, Flv = flow parameter, ∆P/H = pressure loss)
 45

∆
∗ ∗ 1− ∗
=
(1)
∆
1+ ∗ ∗

Table 2: Constants for numerical version of GPDC

GPDC type C1 C2 C3 C4 C5 C6 C7
structured packings
3.8617 0.6609 6.3763 0.7206 0.2898 -0.9093 -0.6819
[in H2O/ft]
random packings
2.7561 0.5778 5.3597 0.5545 0.4046 -1.4234 -0.6022
[in H2O/ft]

In 1991, Robbins developed a relatively simple correlation for the pressure loss prediction based on the
equations developed by Leva 1954 as an alternative to the generalized pressure drop correlation (GPDC)
charts. The equations are simple as no iteration is necessary and the only necessary empirical constant is
called packing factor and can be calculated based on the dry pressure drop results of a packing. The model
uses imperial units instead of SI units so the user has to be careful.
The research group of Piché, Larachi and Grandjean developed a fluid dynamic model based on a neuronal
network analysis. Therefore, they defined 20 dimensionless quantities that were used as basis for the neural
network and based on those they tried hundreds of different compositions of the quantities for the best results.
The model is complex and their group provides an online spreadsheet to calculate the hydraulic properties.
Furthermore, the spreadsheet includes the packing properties for a huge amount of packings as well as the
additionally required packing sphericity. ([Link]/flarachi/)
Spiegel and Meier 1992 developed a model for the prediction of pressure drop and flooding point for four
different structured packings. They divided the pressure drop calculation into a preloading and a loading
region where the pressure drop calculation in the loading region needs the flooding point. The flooding point
can be derived by using the given Wallis diagram. In this chart only values for the four Mellapak packings
Mellapak 125X, 250X, 250Y and 500Y are depicted and therefore the model performance can just be
evaluated for these packings.
Suess and Spiegel 1992 developed a liquid hold-up correlation for the preloading region for structured
packings that is only dependent on the liquid load, the liquid viscosity and the geometrical surface area.
Jammula developed a correlation for liquid hold-up and flooding point for structured packings that does not
need any empirical parameter. The liquid hold-up correlation is based on a modified model for the liquid film
thickness and the correlation for the flooding point gas velocity on the Wallis equation.

4. Modelling results
4.1 Pressure drop modelling results
In general, the examined pressure drop correlations predict the packing pressure drop reasonably well as
shown in Tables 3 and 4. The Robbins correlation shows large deviations for all packings that have a packing
factor FP smaller than 7. The reason for this behavior needs further investigation. The Kister GPDC correlation
shows reliable results for all examined structured packings but for the random packings RSR #0.3, RSR #0.5
and RMSR 25-3, the model predicts too high pressure drops in the loading region which leads also to greater
mean deviations. The NNA model of Piché et al. predicts the pressure drop for most of the examined packings
within a mean deviation of 50% but for some, it even leads to mean deviations of more than 100%. Therefore,
this model should be used with caution. The Spiegel and Meier model provides reliable predictions for the
packings it was developed for but the required use of the Wallis plot is impractical. Good pressure drop
predictions can be achieved with the Brunazzi model, although it needs four empirical constants.
Nevertheless, it is unable to predict distillation pressure drop at higher operating pressures as the evaluation
at 1.6 and 4.1 bar shows.
4.2 Liquid hold-up modelling results
The herein examined models (Tables 5 and 6) are not able to reliably predict the liquid hold-up as for
example, the modified Maćkowiak model or the Billet and Schultes model in Wolf-Zoellner et al. 2018 do.
Nevertheless, it is interesting that with the simple equation for the preloading region of Suess and Spiegel,
originally developed for structured packings, promising results for structured as well as random packings can
be achieved. The Brunazzi model in combination with the Maćkowiak model for the preloading hold-up
46

achieves very good agreement except for systems with a high liquid viscosity (14 mPa*s through adding
polyethylene oxide to water). The huge advantage of the Jammula model is that it does not need any empirical
constants but the results are unsatisfactory.

Table 3: Mean deviation of pressure loss [%] – Air/water system – random packings (* M=Metal, P=Plastic)
NNA-Model
Packing Material* Robbins Kister GPDC
Piché
CMR 2 M 24 39 38
CMR 2A P 9 15 38
IMTP 40 M 11 35 47
1 in Pall Ring M 16 44 49
1 in Pall Ring P 29 45 18
2 in Pall Ring M 7 18 21
RSR #0.3 M 15 49 33
RSR #0.5 M 18 30 26
RSR #0.7 M 19 15 146
Hiflow 50-0 P 37 22 120
Hiflow 50-6 P 96 28 149
Raflux 25-5 M 18 34 19
Raflux 50-5 M 14 29 18
RMSR 25-3 M 11 61 37
RMSR 50-4 M 15 24 110
RMSR 70-5 M 106 23 54

Table 4: Mean deviation of pressure loss [%] – Air/water system – structured packings (* M=Metal, P=Plastic)
Brunazzi Spiegel and
Packing Material* Robbins Kister GPDC
(+ Maćkowiak) Meier
Mellagrid 64Y M 101 21 27 -
Mellapak 125Y M 27 30 23 -
Mellapak 200Y M 21 26 17 -
Mellapak 200X Plus M 74 25 38 -
Mellapak 250X M 36 25 23 20
Mellapak 250X
M 36 30 36 28
high viscosity (14mPa*s)
Mellapak 250Y M 17 24 20 20
RMP N 250Y
M 18 34 125 22
organic/distillation
Mellapak 250Y
M 23 26 16 21
smooth
Mellapak 252Y M 19 17 17 -
Mellapak 252Y
M 19 30 23 -
high viscosity (14mPa*s)
GTC 350Y M 36 28 38 -
RMP S 350Y M 18 28 16 -
Mellapak 500Y M 37 25 17 34
RSP 200 M 19 24 - -
RSP 250 M 17 20 - -
RSP 300 M 20 16 - -
Hiflow Plus #1 P 20 19 - -
Hiflow Plus #2 P 130 22 - -

4.3 Flooding point modelling results

Flooding points can be calculated with the Kister GPDC-model for random and structured packings and with
the Jammula-model for structured packings. The Jammula calculation does not need any empirical packing
constant and predicts all available flooding points for sheet structured packings within a mean deviation of
22%. Based on the Kister correlation, the flooding points for random packings can be predicted within a mean
deviation of 17,5% and for structured packings within a mean deviation of 12%.
 47

Table 5: Mean deviation of hold-up [%] – Air/water system – random packings (* M=Metal, P=Plastic)
Suess und NNA-Model
Packing Material*
Spiegel Piché
1 in Pall Ring P 28 36
RSR #0.3 M 36 75
RSR #0.5 M 47 16
RSR #0.7 M 23 103
Hiflow 50-0 P 51 48
Hiflow 50-6 P 48 40
Raflux 25-5 M 43 15
Raflux 50-5 M 37 56
RMSR 50-4 M 36 62
RMSR 70-5 M 43 182

Table 6: Mean deviation of hold-up [%] – Air/water system – structured packings (* M=Metal, P=Plastic)
Brunazzi Suess and
Packing Material* Jammula
(+ Maćkowiak) Spiegel
Mellagrid 64Y M 30 43 39
Mellapak 125Y M 41 70 35
Mellapak 200Y M 21 40 30
Mellapak 200X Plus M 10 28 17
Mellapak 250X M 25 46 61
Mellapak 250X
M 62 62 77
high viscosity (14mPa*s)
Mellapak 250Y M 17 20 27
Mellapak 250Y
M 25 22 41
smooth
Mellapak 252Y M 21 45 59
Mellapak 252Y
M 72 60 109
high viscosity (14mPa*s)
GTC 350Y M 12 14 16
RSP 200 M - 30 -
RSP 250 M - 39 -
Hiflow Plus #1 P - 19 -
Hiflow Plus #2 P - 30 -

5. Conclusion
Combining the results of the current work and the performance comparison in Wolf-Zoellner et al. 2018, the
modified Maćkowiak-model is highly recommended for fluid dynamic calculations due to its overall good data
agreement for random and structured packings for pressure drop, hold-up and flooding point prediction and
the need of just one empirical constant. For pressure drop and flooding point estimates, the Kister GPDC-
chart is also a good alternative as it is simple and clear in its use, besides its good data agreement.

6. Outlook
Six correlations have been examined within this paper and eight correlations in Wolf-Zoellner et al. 2018 but
there are still further hydraulic correlations in the literature (e.g. Mersmann 1994, Grabbert and Bonitz 1998)
that also should be examined concerning their modelling performance. Furthermore, the available database
should be continuously extended. In particular, it should include a larger variety of test systems. The end
result of this research should be exact recommendations on which model should be used for which packing
type and under which conditions. Therefore, the exact limitations of all models should be examined and
compared with each other.

Acknowledgments

The author would like to thank the Separations Research Program (SRP) Staff for their great support and the
possibility to work with their large pressure drop and hold-up database as well as for the great hospitality
during my six months stay at SRP.
48

References
Billet R., Schultes M., 1991, Modelling of pressure drop in packed columns, Chemical Engineering &
Technology, 14 (2), 89–95.
Billet R., Schultes M., 1999, Prediction of mass transfer columns with dumped and arranged packings,
Chemical Engineering Research and Design, 77 (6), 498–504.
Bozzano G., Dente M., Corna P., 2007, A phenomenological model for fluid-dynamics evaluation of structured
packing systems, AIDIC Chemical engineering transactions, Vol. 12, 189–194.
Brunazzi E., Paglianti A., 1997, Mechanistic pressure drop model for columns containing structured packings,
AIChE J., 43 (2), 317–327.
Brunazzi E., Paglianti A., Spiegel L., Tolaini F., 2002, Hydrodynamics of a gas-liquid column equipped with
MellapakPlus packing. IChemE (Hg.): Proceedings of the International Conference on Distillation and
Absorption, Baden-Baden, Germany.
Engel V., 1999, Fluiddynamik in Packungskolonnen für Gas-Flüssig-Systeme, PhD Thesis, TU Munich,
Lehrstuhl für Fluidverfahrenstechnik, Munich, Germany. (in German)
Grabbert G., Bonitz R., 1998, Fluiddynamik bei der Zweiphasenströmung durch Füllkörper- und
Packungskolonnen, Freiberger Forschungshefte, TU Bergakademie, Freiberg, Germany. (in German)
Gualito J. J., Cerino F. J., Cardenas J. C., Rocha J. A., 1997, Design method for distillation columns filled with
metallic, ceramic, or plastic structured packings, Ind. Eng. Chem. Res., 36 (5), 1747–1757.
Jammula A. K., 2014, New liquid holdup, load point and flooding velocity models in different regions of
operations for a structured packed column, PhD Thesis, Oklahoma State University, Oklahoma, USA.
Kister H. Z., 1992, Distillation design, McGraw Hill, New York, USA.
Kister H. Z., Scherffius, J., Afshar, K., Abkar, E., 2007, Realistically predict capacity and pressure drop for
packed columns, Chemical Engineering Progress, 103 (7), 28–38.
Leva M., 1954, Flow through irrigated dumped packings, Chem. Eng. Progress Symp. Ser., 50, 51-56.
Maćkowiak J., 2003, Fluiddynamik von Füllkörpern und Packungen, Grundlagen der Kolonnenauslegung,
Springer, Berlin, Germany. (in German)
Mersmann A, 1994, Druckverlust und Flutpunkt in berieselten Packungen, VDI-Wärmeatlas, 7. Edition, VDI-
Verlag, Germany. (in German)
Olujić Ž., 1997, Development of a complete simulation model for predicting the hydraulic and separation
performance of distillation columns equipped with structured packings, Chemical and Biochemical
Engineering Quarterly 11 (1), 31–46.
Olujić Ž., Seibert A.F, Fair J.R, 2000, Influence of corrugation geometry on the performance of structured
packings: an experimental study, Chemical Engineering and Processing: Process Intensification 39 (4),
335–342.
Piché S., Larachi F., Grandjean B.P.A., 2001a, Improved liquid hold-up correlation for randomly packed
towers, Chemical Engineering Research and Design, 79 (1), 71–80.
Piché S. R., Larachi F., Grandjean B. P. A., 2001b, Improving the prediction of irrigated pressure drop in
packed absorption towers, Can. J. Chem. Eng., 79 (4), 584–594.
Robbins L. A., 1991, Improve pressure-drop prediction with a new correlation, Chemical Engineering
Progress, 87, 87–91.
Song D., 2017, Effect of Liquid Viscosity on Liquid Film Mass Transfer for Packings, PhD Thesis, University of
Texas at Austin, Austin, USA.
Spiegel L., Meier W., 1992, A generalized pressure drop model for structured packings, IChemE (Hg.):
Distillation and Absorption, B85-B94, Birmingham, UK.
Stichlmair J., Bravo J. L., Fair J. R., 1989, General model for prediction of pressure drop and capacity of
countercurrent gas/liquid packed columns, Gas Separation & Purification, 3 (1), 19–28.
Strigle R. F., 1994, Packed tower design and applications. Random and structured packings, Gulf Pub. Co.,
Houston, USA.
Suess P., Spiegel L., 1992, Hold-up of mellapak structured packings, Chemical Engineering and Processing:
Process Intensification, 31 (2), 119–124.
Tsai R. E., 2010, Mass Transfer Area of Structured Packing, PhD Thesis, University of Texas at Austin,
Austin, USA.
Wolf V., Lehner M., Hoffmann K., 2015, Modification of the Billet and Schultes-Model for the hydraulic
modelling of random and structured packing, AIChE Spring Meeting, 144, Austin, USA.
Wolf V., 2014, Entwicklung eines hydraulischen Modells für eine neuartige Gitterstrukturpackung, PhD Thesis,
Montanuniversitaet Leoben, Chair of Process Engineering, Leoben, Austria. (in German)
Wolf-Zoellner V., Seibert F., Lehner M., 2018, Performance comparison of different pressure drop, hold-up
and flooding point correlations for packed columns, Chem. Eng. Res. Des. (under review).