Int.J, MIdt/p/mx J:]ow VoL 6, pp.

217-225
Perlpmmn/Elsevim',I~0. Printedin GreatBri~in

FLOW PATTERN TRANSITION FOR GAS-LIQUID

F L O W IN H O R I Z O N T A L A N D I N C L I N E D P I P E S

COMPARISON OF EXPERIMENTAL DATA WITH THEORY

DVORA BARNEA, OVADIA SHOHAM and YEHUDA TAITEL

Schoolof Engineering,Tel-AvivUniversity,Ramat-Aviv,Israel
and
A. E. DUKLER
Department of Chemical Engineering, University of Houston, Houston, TX 77004, U.S.A.

(Received 15 February 1979; in revisedform 29 October 1979)

Abstract--Experimental measurements of flow patterns for gas-liquid flow in inclined pipes are reported.
The results compare well with a recently published theory for the prediction of flow patterns in horizontal
and inclined pipes (Taitel & Dukler 1976).

INTRODUCTION
The prediction of flow patterns is a central problem in two phase gas-liquid flow in pipes. Design
parameters such as pressure drop and heat and mass transfer are strongly dependent on the
flow pattern. Hence, in order to accomplish a reliable design of gas-liquid systems such as pipe
lines, boilers and condensers, an a priori knowledge of the flow pattern is needed.
Although extensive research on flow patterns has been conducted, most of this research has
been concentrated on either horizontal or vertical flow. For horizontal flow the earliest and
perhapsthe most durable of pattern maps for two phase gas-liquid flow was proposed by Baker
(1954). Many more have been suggested: Kosterin (1949), White & Huntington (1955), Govier &
Omer (1962) and Mandhane et al. (1974). AI-Sheikh et al. (1970) defined a variety of dimension-
less groups and concluded that no two groups characterize all of the transitions and aU of the
data. Taitel & Dukler (1976) proposed a physical model capable of predicting flow regime
transition in horizontal and near horizontal two phase flow. In this work the flow pattern
boundaries were predicted analytically and it was shown that for each inclination angle every
transition boundary can be represented by two dimensionless groups.
Very little work has been published on the effect of inclination on two phase flow in pipes.
Bonnecaze et al. (1971) and Singh & Griflith (1970) studied slug flow in inclined pipes but they
deal with only one flow pattern and with a limited extent of inclination. Gould et al. (1974)
published flow pattern maps for horizontal and vertical flows and for upflow at 45° inclination.
They also presented a model for predicting pressure distribution in two phase flow through
vertical, inclined and curved pipes. Beggs et al. (1973) investigated the effect of pipe inclination
on liquid holdup and pressure loss. The problem of pressure drop in inclined flow was treated
by Savigny (1962). Greskovich (1972) studied holdup in gas-liquid downflow. Oshimowo &
Charles (1974) investigated vertical downflow in bends.
The present work reports new data on flow pattern transitions for gas-liquid flow in inclined
pipes. The flow of air and water through 2.55 and 1.95cm pipes is studied at flow angles as large
as 10°. The analytical model for predicting flow pattern for horizontal and near horizontal pipe
as proposed by Taitel & Dukler (1976) is tested by comparison with the experimental results.

E X P E R I M E N T A L E Q U I P M E N T AND P R O C E D U R E
The experimental apparatus consisted of an air-water supply system containing a water
pump, air and water flow meters and test section in which the two phase flow is visually
observed and the flow pattern recorded.
217
218 DVORABARNEAet al.

Air was supplied at 60 N/cm 2 from a central compressed air line through an oil separator.
The air passed through a regulating valve to maintain a constant pressure and then through gas
rotameters. Air flow rate was controlled by valves upstream of the meters. Liquid was supplied
from a tank and circulated through the system by a centrifugal pump.
Two different inlet sections were used. The first was a "tee" type with water entering on the
run and air on the branch. The second inlet section was a "mixer" type where air was
introduced circumferentially into the pipe through uniformly distributed small holes. Two test
sections were fabricated of Plexiglas tubing with inner diameters of 1.95 and 2.55 cm. The tubes
were each 3 m long supported on a steel frame which could be inclined with the horizontal.
Flow temperature and pressures were measured.
The flow pattern in the test section was determined by visual observation. Careful preli-
minary runs were made for horizontal flow in order to define, classify and categorize the flow
pattern. Some of the transition boundaries were found to be sharp but most of them were
gradual with respect to the flow rate changes. Experiments were carried out with horizontal
pipes for upward inclination angles of 0.25, 0.5, 1, 2, 5 and 10° and for downward angles of I, 2,
5 and l0 °. Inclination angles reported in this work are accurate to _+0.03°.

CHARACTERISTIC FLOW PATTERNS FOR HORIZONTAL ANDINCI,INEDFI,OW

The designation of flow pattern has been based largely on individual interpretation of visual
observation. While some instrumental methods of analysis have been proposed (Hubbard &
Dukler 1%6, Jones & Zuber 1975), these are not simple to use and have thus not found
widespread application. Furthermore, for many purposes of design a simple approximation to
the location of the gas and liquid phases suffices as a description. For these reasons it is
important to specify as precisely as possible the features of the flow used to characterize the
pattern designated, and in this work the following apply:
Stratified (S). Liquid flows at the bottom of the pipe with gas at the top, The interface can
either be smooth (SS) or wavy (SW).
Intermittent (I). In this flow pattern the inventory of liquid in the pipe is non-uniformly
distributed axially. Plugs or slugs of liquid which fill the pipe are separated by gas zones which
contain a stratified liquid layer flowing along the bottom of the pipe. The liquid may be aerated
by small bubbles which are concentrated toward the front of the liquid slug and the top of the
pipe. The intermittent pattern is usually subdivided into slug (SL) and elongated bubble (EB)
flow patterns, but the distinction between them has not been clearly defined in the past. In this
work the elongated bubble pattern is considered the limiting case of slug flow when the liquid
slug is free of entrained gas bubbles. With this criterion the elongated bubble-slug transition is
quite sharp and easy to identify.
Annular (A). The liquid flows as a film around the pipe wall. A liquid film surrounds a core
of high velocity gas which may contain entrained liquid droplets. The film at the bottom is
normally thicker than at the top depending on the flow rate of liquid and gas. At the lowest gas
rates at which transition to annular from slug flow is observed, most of the liquid flows at the
bottom of the pipe. The upper walls are intermittently wet by large aerated, unstable waves
sweeping through the pipe. This is not slug flow which requires a competent bridge of liquid nor
is it fully developed annular flow which requires a stable film over the entire pipe perimeter.
Thus, it is designated as the wavy annular pattern (WA). Nicholson, Aziz and Gregory (1978)
also recognized the existence of this hybrid pattern between annular and slug flow and
designated it as "proto-slug" flow.
Dispersed bubble (DB). The gas phase is distributed as discrete bubbles within a con-
tinuous liquid phase. In these experiments transition to this pattern is defined by the condition
where bubbles are first suspended in the liquid or elongated bubbles which contact the top of
the pipe are destroyed. When this first happens most bubbles are located near the top. At higher
liquid rates these bubbles are dispersed more uniformly,
 FLOWPATTERNTRANSITIONFORGAS-LIQUIDFLOW 219

RESULTS AND DISCUSSION

Horizontal test section. The results obtained in this experiment are compared in figure 1
with the coordination of experimental data for small tubes represented by the Mandhane map.
Very good agreement exists with the other data used to build Mandhane's map for the basic
transition from the stratified pattern to either the intermittent or annular pattern. These
transitions take place over a narrow change in flow rates, are quite distinct and are easily
reproduced by different investigators.
The Mandhane dispersed bubble-intermittent flow boundary is located at higher liquid rates
than these new data. The difference results from differences in definition. Mandhane et al.
(1974) distinguished elongated bubble from bubble flow only when the bubbles were distributed
more or less uniformly throughout the liquid. From a modelling point of view,
it seems more logical to consider the flow pattern as being dispersed bubble once small bubbles
are observed in a continuous liquid and the elongated bubbles disappear. As a result, our
dispersed bubble regime transition appears at a lower liquid flow rate compared to Mandhane's
map. In the flow regime maps presented by Baker (1954), Hoogendoorn (1959) and Gorier &
Aziz (1972), the dispersed bubble-intermittent transition line is located at lower liquid rates than
in Mandhane's map.
The extent of the annular flow pattern region also differs from that presented by Mandhane.
A closer observation shows that our transition boundary agrees well with Mandhane's if the
wavy annular regime (the black triangular points in figure 1) is considered slug flow. As
discussed above, wavy annular flow has been interpreted as slug flow by some previous
investigators. Thus, this discrepancy is based on a difference in interpretation of the nature of
wavy-annular flow as discussed above.
The slug-elongated bubble transition boundary has not been well defined. Mandhane
followed the description of Gorier & Aziz (1972) in which the distinction between elongated
bubble and slug flow is based on "bubble shape" and film thickness below the bubble. Figure 1
shows that the criteria of no gas holdup in the liquid slug is consistent with previous data for

IO
>:
p-
O
o
..J I.
uJ
>
E3
o
.-I O.I ~
<_
_o o~ oo ~o
b.
,,=, 0.01
¢/)
0.002
0.O2 O.I I IO IOO 2 0 0
SUPERFICIAL GAS VELOCITY, m/s

STRATIFIED SMOOTH (SS}'~

• STRATIFIED WAVY (SW),,~ STRATIFIED (S)
o ELONGATED BUBBLE (EB)'~
• SLUG (SL)J INTERMITTENT(I)
z~ ANNULAR ( A)'~
• WAVY ANNULAR (AW)J ANNULAR (A)
• DISPERSED BUBBLE (DB)
Figure 1. Comparison of data for 2.5 cm horizontal tube with Mandhane et al. (1974) Map and TaiteI-Dulder
(1976) theory.////Mandhane; - - Experiment; - - Theory.
220 DVORABARNEAet al.

elongated bubble-slug transition in spite of the vague definition. Indeed, the experimental
elongated bubble to slug transition compares well with Mandhane's map.
Effect of inclination on flow regime. The effect of upward and downward inclination on
flow pattern transition was carefully examined. Some typical results are shown in figures 2-7.
The data shown were obtained in the 2.55-cm pipe with similar results observed for the 1.95-cm
pipe. The effect of the entrance section was found negligible for all flow pattern boundaries
except for the stratified smooth-stratified wavy transition. In figures 2-7 results for the "mixer"
type entrance section are shown.
The major effect of inclination on the flow patterns is observed in the transition between
stratified and intermittent or annular regimes. In downflow (figures 5-7) the liquid moves more
rapidly and has a lower level in the pipe owing to downward gravity forces. As a result, higher
gas and liquid flow rates are required to cause transition from stratified flow and the stratified
flow region is considerably expanded as the angle of inclination increases. Conversely, upward
inclination cause intermittent flow to take place over a much wider range of flow conditions
(figures 2-4). The stratified-intermittent transition boundary is very sensitive to the angle of
inclination and even for upward angles of less than 1° the regime of stratified flow shrinks into a
small bell-shaped region. For upward inclinations higher than 10° stratified flow is not observed
at all (for the observed range of liquid and gas flow rates).
Stratified smooth flow is not seen for upward inclination except for very small angles (less
than 0.25°). For downflow, contrary to the case of horizontal and upflow where waves are
generated from the action of the "wind" on the interface, natural instability of the interface
occurs as a result of flow due to gravity even in the absence of gas flow. Thus, for negligible gas
flow rates the transition from smooth to wavy interfaces depends only on the liquid rate as
clearly indicated by the "horizontal" transition line which is independent of the gas flow rate.
Unlike the stratified smooth-stratified wavy transition for horizontal flow, where the change is
quite sharp, in downflow the location of this transition cannot be accurately determined. In
addition, changes from stratified smooth to wavy depend on the type of entrance section being
used. The tee-type entry section resulted in a transition boundary at a liquid flow ratio 2-4 times

2O

DO

0 • -_...z~l o • o , I • ~"
J
>
o o o oil o o o • i lP'O,.,,,-
00( ooool • oo oJ¢~
00( • OBOe Ooo~r~
5 00¢ • OBOe ioozl'~ ~
o
J .I
.J A

O0 , o o ooqn • ~ilJ u q l ~ ~
O0 ,oooo ilia• ~
,OI o u c 'L,) U UI II I
OOC 0 0 0 II il qlli
lll
i
,O2 .I I IO 1(30
SUPERFICIAL GAS VELOCITY, m / s

E~ STRATIFIED SMOOTH (SS)'~

• STRATIFIED WARY (SW)J STRAT{FIED (S)
O ELONGATED BUBBLE (EB)'~
• SLUG (SL)J INTERMITTENT ( I )

ANNULAR (A) "~ ANNULAR (A)

• WAVY ANNULAR (AW)J
• DISPERSED BUBBLE (DB)

Figure 2. Experiment vs theory. 0.25 ° upward inclination of a 2.5 cm pipe. --, experiment: ---, theory,
 FLOW PATTERN TRANSITION FOR GAS=LIQUID FLOW 221

20
E I0

I,-

0
.~,~~.,-'"
-,j O°ooooo
oo.. if:=,i
oo.
O O• •••~
"0.1 O O. ;..~.
.J o Ioo e• i A
o
o 0 • • • O|•I~
oo<
O O • • A
"' 0.01 u

(/') 0 0 • • •
O0 000 • @ill
0.002
0,02 O.I I IO IOO 200
SUPERFICIAL GAS VELOCITY, m / s

a STRATIFIED SMOOTH (SS)'~

• STRATIFIED W A V Y ( s w ) J STRATIFIED (S)

o ELONGATED BUBBLE (EB) l INTERMITTENT ( I )

• SLUG (SL) j
A ANNULAR ( A ) '~
• WAVY ANNULAR (AW)j ANNULAR (A)
• DISPERSED BUBBLE (OB)

Figure 3. Experiment vs theory. 20 upward inclination of a 2.5 cm pipe. - - , experiment, ...... , theory.

20

E DB

.j
,. - - -O~' o-O~O O~ ,i . I I IIi~ /
I ,>"
i,i O ,0000 tl• /
o .ooo•••••
o o o o o . . . . . _
5 o I De •e :6~ I
_0 0.I ~ ui l,i I
,_1
o De •DO•/ I A
_1 o DO •o••z I
¢I o ooooioo•&z I~.
o DO o•o •-'
u. O.OI o o. .I;i•~
I=.1 OC O• • &~
8.
-,j O oooeeeee•~l~A
o~ 0 0 0 2 i
002 OI I IO I00 200
SUPERFICIAL GAS VELOCITY, m/s

n STRATIFIED SMOOTH (SS) l

• STRATIRED W A V Y (SW)~ STRATIFIED (S)
o ELONGATED BUBBLE (EB)'~
• SLUG ( S L ) J INTERMITTENT ( I )
z~ ANNULAR (A)
• WAVY ANNULAR (AW)J ANNULAR (A)
• DISPERSED BUBBLE (DB)

Figure 4. Experimental vs theory. I0 ° upward inclination of a 2.5 cm pipe. - - , experiment; ..... , theory.

MF VoL 6,No, 3---=D

222 DVORA BARNEA el ai

20
I0
E

0
_J
ILl

0
QI
_J

la..
001
W

o3
0.002
0.02 I I I0 I00 200
SUPERFICIAL GAS VELOCITY,m/s

D STRATIFIED SMOOTH (SS)

• STRATIFIED WAVY (SW) J STRATIFIED (S)
o ELONGATED BUBBLE (EB)"~ INTERMITTENT(I)
• SLUG (SL) J
,~, ANNULAR (A) "~
• WAVYANNULAR (AW)J ANNULAR (A)
• DISPERSED BUBBLE (DB)

Figure 5. Experimental vs theory. 1° d o w n w a r d inclination of a 2.5 c m pipe. - - , experiment; ....... , theory.

2O
IO
E DO ._-=- ~__--.j
>_-

o, I.
Tj
hi
>

E~
5o 0.1
.J

G
001
r,-
hi

(n 0 . 0 0 2
.OZ .I I Io IO0 200
SUPERFICIAL GAS VELOCITY, m/s

c STRATIFIED SMOOTH (SS)-~

STRATIFIED (S)
• STRATIFIED WAVY (SW) f
© ELONGATED BUBBLE (EB) 1
• SLUG (SL) yJ INTERMITTENT (I)
ANNULAR ( A ) "1
• WAVY ANNULAR (AW) ~ ANNULAR (A)
• DISPERSED BUBBLE (DB)

Figure 6. Experimental vs theory. 5 ° d o w n w a r d inclination of a 2.5 cm pipe. - - , experiment; ...... . theory.

 FLOW PATTERN TRANSITION FOR GAS-LIQUID FLOW 223

20
I0
E DB
i//s
O • O O O /
O
I
I. ii i i i i i
I~I • • • m4r,,~• • oo.....~
• • • mJn~A
• • •
• •
_o 0.1
A
--J • HI

" "- i- s III

,u,,~
I Z~

0.01
... ,,-_
--m
!,,
i J
I • I I l
r-I 1~ rn n •

ooo2i
0.02 0.I I0 I00 200
SUPERFICIAL GAS VELOCITY, m/s

[] STRATIFIED SMOOTH ( S S ) ~ STRATIFIED (S)

• STRATIFIED WAVY (SW) J
o ELONGATED BUBBLE (EB) "L
INTERMITTENT ( I )
• SLUG (SL) J
z~ ANNULAR (A) "~ ANNULAR (A)
r
• WAVY ANNULAR (AW).J
• DISPERSED BUBBLE (DB)
Figure 7. Experimental vs theory. 10° downward inclination of a 2.5cm pipe. --, experiment; ...... , theory.

higher than that for the mixer-type section. Because of the difliculty in accurately determining
this transition, the results should be treated with caution.
The experimental data show that for upward inclinations the flow rate space over which the
intermittent pattern exists expands at the expense of the stratified pattern while for inclined
downflow the result is just the opposite. An isolated region of stratified flow occurs so that as
the gas rate is decreased, flow pattern transitions from annular to stratified and then back to
annular or slug flow can be observed (see figure 3). This unexpected result is confirmed both
from experiment and theory.
Comparison of the theoretical model with experimental data. The results of these experi-
ments are compared with the theoretical model previously published by Taitel & Dukler (1976)
in figures 2-7. This theory was designed to predict the flow pattern transitions for horizontal
and "near horizontal" flows. At the time the theoretical model was conceived, experimental
verification was possible only for the horizontal case because of the lack of data for inclined
pipes. Now it is possible to check the validity of this model for inclined pipes as well as
determine quantitatively the range of inclination angles for which the theory can be considered
valid.
The broken lines in figures 1-7 represent the theoretically predicted transition boundaries.
The agreement between experiment and theory is indeed remarkable up to _10 ° inclinations,
except for the stratified smooth to wavy transition for downward inclinations.
For horizontal flow (figure 1) very satisfactory agreement exists with respect to all transition
lines. The apparent discrepancies are of the same order as the experimental error.
The comparisons displayed in figures 2-7 lead to the conclusion that the model presented by
Taitel & Dukler (1976) is applicable to inclined pipes to _+10° inclination with good accuracy.
The agreement is particularly well demonstrated for inclined upflow (see figures 2--4) where the
bell-shaped stratified region is quite accurately predicted. Experimental measurements not
224 DVORABARNEAet al.

displayed here show that for angles from 10 to 30°, the theories while not accurate can give a
reasonable estimate of the pattern.
The only exception to this agreement is the transition boundary between stratified smooth
and stratified wavy patterns especially for downward flow. The theoretical model fails to
correctly account for the start of the wavy stratified pattern for downward flows. According to
the model, the waves are caused by the gas flow as a result of transfer of energy between the
gas and liquid phase. However, as has been observed waves will develop on the surface of a
falling film in the absence of gas flow. The existence of these waves depends on the liquid flow
rate and the level of the interface and this changes significantly with inclination. Because the
disagreement is so substantial, the theoretical results which are based on Taitel & Dukler (1976)
for this transition are not displayed on the graphs to avoid confusion.
Theories for predicting conditions for wave formation on films flowing on inclined planes
have been presented by Benjamin (1957), Yih (1%9), Stoker (1957), Krantz & Goren (1970),
Whittaker (1966), Hanratty & Hershman (1%1) and many others.
Depending on the particular model used, the criteria for the first appearance of waves is
described in terms of Reynolds number, Froude number and sometimes the Weber number. The
usual approach is to perturb the Navier-Stokes equations leading to an Orr-Sommerfield type
equation which is solved either numerically or by various approximate means to determine the
condition for stability. A fairly comprehensive review of existing solutions is given by Fulford
(1%4) and Krantz & Goren (1971).
Although results of the various investigators differ from each other, the general agreement is
that the interface is unstable at a very low Reynolds number of the order of 5-50 depending on
the angle of inclination. As a typical example, consider the classical treatment by Yih (t%9)
who showed that the critical Reynolds number for the generation waves is given by

Rc = 6 c o t c~. Ill

According to this result the interface is always unstable for the vertical plane (in practice, it is
unstable for very low Reynolds numbers) and always stable for the horizontal case (a = 0),
However, for inclination over 1°, the critical Reynolds number is less than 50. Comparison with
these experiments shows that even for the mixer-type entrance element, smooth interfaces
were observed at much higher Reynolds numbers. Also, the trend as given by Ill does not agree
with observation.
Its suggested the fact that smooth stratified surfaces were obtained at flow rates where the
theoies predict a wavy interface may be due to entrance phenomena and the low rate of
amplification expected at low liquid rates (Benjamin 1957). This hypothesis may be supported
by the fact that in the "tee" type entrance section the interface was smooth at liquid flow rates
two to four times higher than for the "mixer" entrance section.

SUMMARY AND CONCLUSIONS

Experimental results were collected for a water-air system in 2.55- and 1.95-cm pipes and
compared with the theoretical prediction of Taitel & Dukler (1976) up to _+10° inclination. Over
this range the agreement between experiment and theory has been found to be very good; At
larger inclinations the trends predicted by the theory are correct but quantitative predictions are
inaccurate.
These results are considered as experimental proof of the validity of the theory that until
now has been tested only for horizontal pipes. The theory fails to predict the stratified-smooth
to stratified wavy transition for inclined pipes. This is significant only for downflow.
 FLOWPATTERNTRANSITIONGAS-LIQUID
FLOW 225

Acknowledgement--This work was sponsored by the U.S.--Israel Binational Science Foundation.

The support of this foundation is greatly acknowledged.

REFERENCES
AL-SHEIKH, J. N., SAUNDERS,D. E. & BRODKEY,R. S. 1970 Prediction of flow patterns in
horizontal two-phase pipe flow. Can. J. Chem. Engng 48, 21-29.
BAKER,O. 1954 Simultaneous flow of oil and gas. Oil and Gas J. 53, 185-195.
BEGGS,H. D. & BRILL,J. P. 1973 A study of two phase flow in inclined pipes. J. Pet. Tech. 25,
607--617.
BENJAMIN,T. B. 1957 Wave formation in laminar flow down an inclined plane. J. Fluid Mech. 2,
554-574.
BONNECAZE,R. H., ERSKINE,W. & GRESKOVlCH,E. J. 1971 Holdup and pressure drop for two
phase slug flow in inclined pipelines. AIChEJ. 17, 1109-1113.
FULFORD,C. D. 1964 The flow of liquid in thin films. J. Adv. Chem, Engng 5, 151-236.
GOULD, T. L., TEK, M. R. & KATZ,D. L. 1974 Two phase flow through vertical, inclined or
curved pipes. J. Pet. Tech. 26, 915-926.
GOVIER,G. W. & OMER,M. M. 1962 The horizontal pipeline flow of air-water mixtures. Can. J.
Chem. Engng 40, 93-104.
GRESKOVlCH,E. J. 1972 Holdup prediction for stratified downflow of gas-liquid mixtures. Ind.
Engng Chem., Process Design Devel. 11, 81-85.
HASRAaTV,T. J. & HERSHMAS,A. 1961 Initiation of roll waves. AIChEJ. 7, 488-497.
HEwlrr, G. F. & HALL-TAYLOR,N. S. 1970 Annular Two-Phase Flow. Pergamon Press, Oxford.
HOOGENDOORN,C. J. 1959 Gas liquid flow in horizontal pipes. Chem. Engng Sci. 9, 205-217.
HUBBARD,M. G. & I~KLER,A. E. 1966 Characterization of flow regimes for horizontal two-phase
flow: statistical analysis of wall pressure fluctuations. Proc. 1966 Heat Transfer and Fluid
Mechanics Con[., Santa Clara, Calif.
JONES, O. C. & ZUBER,N. 1975 The interrelation between void fraction fluctuations and flow
patterns in two-phase flow. Int. J. Multphase Flow 2, 273-306.
KOSTERIN,S. I. 1949 An investigation of the influence of diameter and inclination of a tube on
the hydraulic resistance and flow structure of gas liquid mixtures. Izvest. Akad. Nauk, SSR
12, 1824.
KRANTZ,W. B. & GOREN,S. L. 1970 Finite-amplitude, long waves on liquid films flowing down a
plane. Ind. Engng Chem. Fund. 9, 107-113.
KRANTZ,W. B. & GORES, S. L. 1971 Stability of thin liquid films flowing down a plane. Ind.
Engng Chem. Fundl. 10, 91-101.
MANDI-IANE,J. M., GREGORY,G. A. and AZlZ,K. 1974 A flow pattern map for gas liquid flow in
horizontal pipes. Int. J. Multiphase Flow l, 537-553.
Molssls, R. 1963 The transition form slug to homogeneous two-phase flows. J. Heat Transfer
366-370.
NICI-IOLSON,M. K., AZlZ, K. and GREGORY,G. A. 1977 Intermittent two phase flow: predictive
models. 27th Can. Chem. Engng Conf., Calgary, Alberta.
NICKLIN,D. J. & DAVlDSON,J. F. 1962 The onset of instability of two-phase slug flows. Instit.
Mech. Engng, Proc. Syrup. on Two-Phase Flow, Paper 4.
SEWGNY,R. 1962 Investigation of isothermal co-current two fluid, two-phase flow in an inclined
tube. Ph.D. Thesis, University of Rochester, Rochester, New York.
SINGH,G. & GRt~TH, P. 1970 Determination of the pressure drop and optimum pipe size for a
two-phase slug flow in an inclined pipe. TRASME, J. Engng Ind. 92, 717--726.
STOKER,J. J. 1957 Water Waves. Interscience, New York.
TAITEL,Y. & DUKLER,A. E. 1976 A model for predicting flow regime transition in horinzontal and
near horizontal gas liquid flow. AIChE J. 22, 47-55.