International Journal of Multiphase Flow 37 (2011) 535–536

Contents lists available at ScienceDirect

International Journal of Multiphase Flow

j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / i j m u l fl o w

Letter to the Editor

Comment on ‘‘Correlation of entrainment for annular flow in horizontal pipes’’,

by Pan, L., Hanratty, T.J., Int. J. Multiphase flow, 28(3), (2002), pp. 385–408
A. Al-Sarkhi ⇑,a, C. Sarica b
a
King Fahd University of Petroleum & Minerals, Dhahran, Saudi Arabia
b
The University of Tulsa, Tulsa, OK, USA

a r t i c l e i n f o a b s t r a c t

Article history: The above referenced paper, published in International Journal of Multiphase Flow (Pan and Hanratty,
Available online 24 January 2011 2002), proposed an entrainment fraction correlation for annular flow in horizontal pipes. The entrain-
ment fraction in annular flow is defined as the ratio of the mass flow rate of the liquid droplets in the
Keywords: gas to the total mass flow of liquid, FE = WLE/WL. The proposed correlation was verified with experimental
Gas–liquid pipe flow data for liquids with viscosities close to that of water. The proposed entrainment fraction correlation
Annular flow includes another correlation for the critical film flow rate, WF,cr to estimate a maximum entrainment frac-
Entrainment
tion FE,max. It is shown that the critical film flow rate correlation can result in negative maximum entrain-
Horizontal pipe
ment fraction values, for low liquid flow rates.
Ó 2011 Elsevier Ltd. All rights reserved.

1. Review Where, WL is the total mass flow of liquid, and WF,cr is the critical
liquid film flow rate below which atomization does not occur. WF,cr
Pan and Hanratty (2002) developed an entrainment fraction is calculated using a correlation based on experimental work by
correlation for low viscosity fluids based on a balance between Andreussi et al. (1985) given by Eq. (3).
the rates of droplet deposition and atomization. For low viscosity
W F;cr ¼ 0:25lL pDReF;cr : ð3Þ
liquids, the initiation of atomization occurs when disturbance
waves appear on the liquid layer. Effects of gravity and droplet size ReF,cr is expressed as a sole function of fluid properties through a
on entrainment are considered. Pan and Hanratty correlations are parameter w. ReF,cr and w are respectively given by Eqs. (4) and (5).
the only correlations to explicitly consider a critical film flow rate
in the maximum entrainment fraction calculation. The entrain- ReF;cr ¼ 7:3ðlog wÞ3 þ 44:2ðlog wÞ2  263 log w þ 439: ð4Þ
ment correlation is defined as qffiffiffiffiffiffiffiffiffiffiffiffiffi
 3 pffiffiffiffiffiffiffiffiffiffiffi !1=ð2mÞ W ¼ ðlL =lG Þ qG =qL : ð5Þ
F E =F E;max 8 DmG qL qG qG1m lmG
¼ 9  10 : ð1Þ
1  F E =F E;max r 1þm
d32 g qL where lL is the liquid viscosity.
Pan and Hanratty (2002) compared their correlation predictions
Where FE is the entrainment fraction, FE,max is the maximum with the experimental data of Dallman (1978), Laurinat (1982),
entrainment, vG is the gas velocity, D is the pipe diameter, d32, is Williams (1990), and Paras and Karabelas (1991). They reported
the Sauter mean diameter of droplets, m is the drag related param- good agreement with data for the 23.1-mm pipe ID, under predic-
eter, qL and qG are liquid and gas densities, lG is the gas viscos- tion for the 50.8-mm ID data, and over prediction for the 95.3-mm
ity.Values of 0, 0.6, or 1 are used for the exponent m in Eq. (1) ID data. Eq. (4) was derived from data obtained for w values from
depending on the drag coefficient calculation method of the termi- 1.8 to 28, and it should not be used outside this range (Pan and
nal velocity of droplets. The Sauter mean diameter is calculated Hanratty (2002).
using one of two correlations presented in the work.FE,max is
expressed with Eq. (2).
2. Discussion
F E;max ¼ 1  W F;cr =W L : ð2Þ
Many cases with low liquid flow or low superficial liquid Rey-
nolds number, ReSL = qLvSLD/lL, results in WF,cr to be greater than
⇑ Corresponding author. Tel.: +966 3 8607725; fax: +966 3 9602949. WL leading to a negative value for FE,max as shown in Fig. 1. This
E-mail address: [email protected] (A. Al-Sarkhi). is due to the fact that WF,cr is developed as a function of parameter

0301-9322/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.ijmultiphaseflow.2011.01.009
536 A. Al-Sarkhi, C. Sarica / International Journal of Multiphase Flow 37 (2011) 535–536

cavity pump. The experimental results were compared with exist-

ing models and correlations, and the best predicting methods were
determined for all flow orientations.
Mantilla (2008) conducted his experiments in an experimental
facility with a test section 13 m long and internal diameter of
0.0486 m. Gas and liquid flow rates are measured using MicroMo-
tion mass flow meters. The entrainment fraction measured using
the liquid film extractor, located 12.3 m (L/D = 253) downstream
of the inlet.

3. Summary

Eqs. (3)–(5) are valid for certain range of total mass flow of
liquid, WL and should be modified and tested at low and high range
of liquid mass flow rate. The data of Magrini (2009) and Magrini
Fig. 1. Maximum fraction of entrainment modeled by Pan and Hanratty vs.
et al. (2010) and Mantilla (2008) and Mantilla et al. (2009) for ReSL
superficial liquid Reynolds number.
less than 350 results in a negative value for maximum entrainment
fraction, FE,max. There is a need to have reliable estimates of WLFC to
w. This parameter is only a function of the physical properties of predict entrainment this requires an accurate and systematic
gas and liquid ignoring the effects of the total liquid flow rate WL entrainment rate and film Reynolds number data at small entrain-
on the WLFC. In other words ignoring liquid flow rate dependency ments and at the onset of entrainment as well as at the maximum
means the disturbance wave effects, which are very strong func- entrainment.
tion of WL, do not exist. This is certainly not correct.
Pan and Hanratty’s approach was tested with air and water data References
from Magrini (2009) and Magrini et al. (2010) and Mantilla (2008)
Andreussi, P., Asali, J., Hanratty, T.J., 1985. Initiation of roll waves in gas–liquid
and Mantilla et al. (2009) with a range of w between 1.97 and 3.1. flows. AIChE J. 31, 119–126.
Although the w values were within the validity range of Eq. (4), Dallman, J.C., 1978. Investigation of Separated Flow Model in Annular Gas-Liquid
negative FE,max values were obtained for the all data with Two-Phase Flow. Ph.D. Dissertation, University of Illinois at Urbana, Champaign.
Laurinat, J.E., 1982. Studies on the Effects of Pipe Size on Horizontal Annular Two-
ReSL < 275. As an example case consider an air–water annular flow Phase Flows. Ph.D. Dissertation, University of Illinois at Urbana, Champaign.
at standard conditions in 0.072 m ID pipe with qL = 1000 kg/m3, Magrini, K., 2009. Liquid entrainment in annular gas–liquid flow in inclined pipes.
lL = 0.001 Pa s, ReFcr = 370, WF,cr = 0.022144 kg/s. Then, FE,max will M.S. Dissertation, The University of Tulsa.
Magrini, K.L. Sarica, C., Al-Sarkhi, A., Zhang, H.-Q., 2010, Liquid Entrainment in
be zero if WL = WF,cr = 0.022144 kg/s. This means all the cases with
Annular Gas/Liquid Flow in Inclined Pipes. 2010 SPE Annual Technical
superficial liquid velocity, vSL < 0.004856 m/s will have negative Conference and Exhibition held in Florence, Tuscany, Italy, 20–22 September,
value of FE,max, Magrini (2009) and Mantilla (2008) reports FE val- SPE paper # 134765.
ues of 0.28–0.45 and 0.09–0.34 respectively corresponding to the Mantilla, I., 2008. Mechanistic Modeling of Liquid Entrainment in Gas in Horizontal
Pipes. Ph.D. Dissertation, University of Tulsa.
negative points shown in Fig. 1. Mantilla, I., Gomez, L., Mohan, R., Shoham, O., Kouba, G., Roberts, R., 2009.
Margini (2009) utilized film removal and iso-kinetic sampling Experimental Investigation Of Liquid Entrainment In Gas In Horizontal Pipes.
techniques to measure the entrainment fraction in a facility with In: Proceedings of ASME 2009 Fluids Engineering Division Summer Meeting,
August 2–5, Vail, Colorado USA, FEDSM2009-78420.
a test section 17.5 m long and 76.2-mm internal diameter. The Pan, L., Hanratty, T.J., 2002. Correlation of entrainment for annular flow in
facility is inclinable from 0o to 90o from horizontal. The entrain- horizontal pipes. Int. J. Multiphase Flow 28, 385–408.
ment fraction measurements were made 13 m downstream of Paras, S.V., Karabelas, A.J., 1991. Droplet entrainment and deposition in horizontal
annular flow. Int. J. Multiphase Flow 17, 455–468.
the inlet (L/D = 172). Air and water were used as the working fluids. Williams, L.R., 1990. Effect of Pipe Diameter on Horizontal Annular Two-Phase Flow.
The compressed air was supplied by dry rotary screw and two- Ph.D. Dissertation, University of Illinois at Urbana, Champaign.
stage compressors. Both gas and liquid were metered using Coriolis
type mass flow meters. The water is pumped using a progressive