Mechanism of heavy oil recovery by cyclic superheated

steam stimulation
Anzhu Xu
a,n
, Longxin Mu
a
, Zifei Fan
a
, Xianghong Wu
a
, Lun Zhao
a
, Bing Bo
a
, Ting Xu
b
a
Research Institute of Petroleum Exploration and Development, PetroChina, Beijing, China
b
Planning Department-Upstream Sector, China National Petroleum Corporation
a r t i c l e i n f o
Article history:
Received 6 August 2013
Received in revised form
25 September 2013
Accepted 26 September 2013
Available online 23 October 2013
Keywords:
heavy oil
superheated steam huff and puff
heat transfer coefcient
degree of superheat
ideal gas
recovery mechanisms
a b s t r a c t
At the conclusion of several cycles of conventional saturated steam huff and puff in heavy oil reservoirs,
the heating radius are typically only 1020 m as it went through successive saturated steam huff and
puff. The heating region cannot be enlarged by continuing saturated steam stimulation any more.
However, superheated steam huff and puff as an additional recovery method signicantly increased the
heating radius by about 10 m after saturated steam huff and puff was completed. Conventional saturated
steam stimulation theory is not applicable for superheated steam. In this study, several physical
simulation tests were performed and reservoirs from conventional saturated steam played different
roles in heavy oil . There exists a temperature gradient on the heat transfer surface, in which chemical
reactions would happen such as formation water, heavy oil, and formation minerals under the favorable
circumstances of high temperature of superheated steam. These chemical reactions not only result in
some changes of composition and the irreversible reduction of oil viscosity, but also change the
microscopic pore structure of rocks to improve the permeability of superheated steam heating area;
changing the wettability of the reservoir rock and increasing the displacement efciency of superheated
steam ooding, which bring about signicant improvements of heavy oil development effects. Due to the
release of latent heat and the uniform temperature of saturated steam, temperature in the saturated
steam heating area is kept constant everywhere. These mechanisms do not occur in ordinary saturated
steam huff and puff. This work analyses the superior properties of superheated steam and bring forward
the superiority of superheated steam huff and puff to effectively develop heavy oil reservoirs in recovery
mechanisms, including mathematichal model establishing, physical and numerical simulation studies,
and current pilot test effects.
& 2013 Elsevier B.V. All rights reserved.
1. Introduction
After several cycles of saturated steam huff and puff in heavy oil
reservoirs, the heating radius are typically only 2030 m, the
maximum radius in a perfect reservoir is about 50 m (Zhang et al.,
2008). In the ultra-heavy oil reservoir, oil saturation stays in the
original state out of the heating radius as it went through successive
saturated steam huff and puff (Wu et al., 2010). The heating region
cannot be enlarged by continuing saturated steam huff and puff any
more due to the limited heat carried by saturated steam and the
serious heatloss during its transmission. In conventional heavy oil
reservoirs, oil has a certain capacity of owing in the original
formation. When the reservoirs went through stages by depletion
or some cycles of saturated steam huff and puff, the reservoir
pressure dropped quickly and water formationwas accelerated to
invade the oil reservoir because of high oil and water mobility ratio.
Water-cut increased rapidly and oil production decreased sharply
after water went through the bottom hole. The production perfor-
mance was becoming worse and worse without the change of
development methods. These reservoirs were inappropriate for
continuing saturated steam huff and puff and their properties
probably hardly met the criteria of saturated steam drive (Qu,
2009). There is a large amount of hydrocarbon accumulation in
such reservoirs that can only be exploited with new concepts. The
secondary enhanced oil recovery technology should be considered
to improve oil production (Yan and Ren, 2009). By heating the
saturated steam above its saturation temperature would be a new
technology for the recovery of these heavy oil reservoirs. Super-
heated steamwas never used to enhance oil recovery before. On the
one hand, superheated steam exists in the very srict conditions and
the superheated state was not directly from cold water heating. On
the other hand, mobile superheated steam generator that can be
used for injectors in the oil eld were not manufactured success-
fully. Since 2005, the rst superheated steam generator has been
Contents lists available at ScienceDirect
journal homepage: www.elsevier.com/locate/petrol
Journal of Petroleum Science and Engineering
0920-4105/$ - see front matter & 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.petrol.2013.09.007
n
Corresponding author. Tel.: 86 108 359 3041; fax: 86 108 359 3245.
E-mail addresses: xuanz@petrochina.com.cn, xuanzhu@cnpcint.com (A. Xu).
Journal of Petroleum Science and Engineering 111 (2013) 197207
put into practice in the Post-salt Oil Field of Kazakhstan. This work
analysis the superior properties of superheated steam and bring
forward superiority of superheated steam huff and puff to effec-
tively develop these marginal heavy oil reservoirs in recovery
mechanisms, including mathematichal model establishing, simula-
tion studies, and current pilot test effects.
2. Properties of superheated steam
Dry saturated steam (quality of 100%) continue to be heated in
the given pressure and the steam temperature will increase above
its saturation temperature. The steam superheated by the number
of temperature degrees higher than the corresponding saturated
steam is dened as superheated steam. The temperature difference
between superheated steam and its saturated steam is called degree
of superheat. As compared to saturated steam under the same
pressure, superheated steam has a higher temperature, carries more
heat and has a greater heating capacity than saturated steam. In
saturated steam, temperature is directly proportional to pressure
and the steam pressure determines the steam temperature,
enthalpy and specic volume; with superheated steam there is no
direct relationship between temperature, pressure and the specic
heat capacity (Rohsenow et al., 1992). Therefore at a particular
pressure it may be possible for superheated steam to exist at a wide
range of temperatures. This is a useful increase in steam energy,
enthalpy and specic volume within the limited pressure (Shen
et al., 2000). With the same quantity of saturated steam, super-
heated steam has the capacity of heating the heavy oil reservoir to a
higher temperature and a wider region. Superheated steam huff
and puff has the potential superiority to improve thermal recovery
effect. Table 1 displays that superheated steam carries more heat
and has larger specic volume than that of saturated steam.
2.1. The conditions for the existence of superheated steam
In thermodynamics, the presence of water can be dened as four
states, such as cold water, wet steam, saturated steam and super-
heated steam. As is shown in Fig. 1. From the region of division,
conditions for the existence of superheated steam are wide ranges
of temperatures and pressures, as is shown in Table 2. Not only at
high temperature and high pressure, superheated steam can also
exist at low temperature and low pressure and even at high
temperature and low pressure. For saturated steam, the tempera-
ture of saturated steam is determined by the pressure (Chang et al.,
1997). But the maximum pressure is limited by steam boiler and
burst pressure of the reservoir rock. Therefore, the saturated steam
temperature increase is limited by the pressure (Li et al., 2008).
But for superheated steam, under a particular pressure, temperature
was higher than the saturated temperature, that is, the state of
steam changed from the saturated region (Region 4) to the super-
heated steam region (Region 2) (Chen et al., 2002). As a result,
superheated steam can not only increase the temperature of steam,
but also increase the steam heat carrying. Superheated steam
breaks through restrictions of saturated steam in application, and
is suitable for thermal recovery of various heavy oil reservoirs.
2.2. Superheated steam generated process
Superheated steam has seldom been reported to be used to
recover heavy oil before due to the limitation of boiler's properties.
However, through two years of research, an innovative model of
Nomenclature
Q heat transferred rate of uids, w
Q
sup
heat transferred rate of superheated steam, w
Q
s
heat transferred rate of saturated steam, w
W friction energy from inlet to outlet of microelement of
wellbore, J/kg
V superheated volume, m
3
T steam temperature, 1C
A heat transfer area, m
2
T
sup
temperature of superheated steam, K
T
s
temperature of saturated steam, K
T

degree of superheat, K
T
w
temperature of heat transfer surface, K
heat transfer coefcient of uids,
W/(m
2
K)

sup
heat transfer coefcient of superheated steam, W/
(m
2
K)

s
heat transfer coefcient of saturated steam, W/(m
2
K)

2
steam density of into and out of microelement of
wellbore, kg/m
3

m
steam density of all microelement of wellbore, kg/m
3
v steam injection rate, m/s
v
1
; v
2
steam injection rate of inlet and outlet of microele-
ment of wellbore, m/s
i
s
steam mass ow rate, kg/s
p steam pressure of microelement of wellbore, Pa
h
m
enthalpy of superheated steam, J/kg
trend angle of deviation
C
p
heat capacity at constant pressure, J/(kg K)
t
s
temperature of saturated steam,1C
p
s
pressure of saturated steam, MPa
z well depth, m
Table 1
Heat and volume multiple of superheated steam compared to saturated steam (quality of 75%).
Pressure Saturation temperature Degree of superheat (10 1C) Degree of superheat (50 1C) Degree of superheat (80 1C)
(MPa) (1C) Heat multiples Volume multiples Heat multiples Volume multiples Heat multiples Volume multiples
1 184.2 1.23 1.37 1.27 1.52 1.30 1.63
3 235.7 1.20 1.37 1.25 1.55 1.29 1.67
5 265.2 1.19 1.38 1.24 1.58 1.28 1.71
7 286.8 1.18 1.38 1.24 1.62 1.28 1.77
9 304.2 1.17 1.39 1.24 1.66 1.28 1.83
10 311.8 1.16 1.40 1.24 1.69 1.28 1.86
A. Xu et al. / Journal of Petroleum Science and Engineering 111 (2013) 197207 198
boiler was successfully built by overcoming many challenges. The
superheated steam rated ow has three options such as 6 t/h,
11.5 t/h and 23 t/h. The injection pressure also has three options
such as 3 MPa, 9 MPa, 19 MPa. The range of superheated degree is
from 30 to 100 1C. The outlet temperature of the superheated
steam generator can reach 320340 1C. This generator is used for
saturated steam with quality of 5070% heated to superheated
steam. The working principle illustrations are shown in Fig. 2. The
wet steam (qualityr50%) must be passed through an additional
heat exchanger to improve its quality (qualityZ90% ), and then
through steam-water separators to detach the water phase. The
detached dry saturated steamwas superheated by passing through
a second heated stage in an additional separate superheater unit of
the boiler.
2.3. Heat transfer characteristics of superheated steam
The intensity of heat transfer is characterized by the wellbore
and formation of heat transfer coefcient. However, the term
intensity of heat transfer does not associate with the heat transfer
coefcient; its denition is that the intensity of heat transfer is
dened as the heat ux per unit area. The physical meaning of
heat transfer coefcient is one unit heat passed through one unit
heat transfer area in one unit time under 1 1C temperature
difference (IAPWS-IF97, 1997). Obviously, the greater the heat
transfer coefcient, the more the heat transfer per unit time. For
the steam carrying a certain quantity of heat, the smaller the heat
transfer coefcient, the less heat ow in the unit time, and the
longer the heat transfer duration. Table 3 lists the heat transfer
coefcients of different uids in heat transfer, which indicates
that, a phase change of uids during the process of heat transfer
will lead to greater heat transfer coefcients, and no phase change
of uids in heat tranfer has smaller heat transfer coefcients; gas
has the smallest transfer coefcient. Fig. 3 demonstrates that
superheated steam exists in the region above the saturation line
(Quality x1) and belongs to 100% degree of dry gas. The trend of
isotherms in Fig.3 shows the higher the degree of superheat, the
closer superheated steam is to ideal gas (Shen et al., 2000; Li et al.,
2008), and smaller the heat transfer coefcient . The calculation
results of heat transfer coefcient shown in Table 3 signify that the
heat transfer coefcient is approximately equal to that of air and is
only 1/1501/250 as much as saturated steam. In addition, super-
heated steam has no phase change in heat transfer and the heat
transfer coefcient is small. But for saturated steam, the loss of
heat will cause some of the steam to condense and phase changes
occur and the heat transfer coefcient is larger. According to
Newton's law of cooling (Eq. (1)), the heat ow rate depends on
the heat transfer coefcient, heat transfer area and the tempera-
ture difference. In the same heat transfer area, the temperature
difference of superheated steam is greater than that of saturated
steam, but the heat transfer coefcient for superheated steam
is always much lower than that of saturated steam and the
temperature difference is relatively much less than the heat
transfer coefcient difference. The higher the degree of superheat,
the lower the heat ow rate. Eq. (2) shows that, in passing along
the same heat transfer area, the heat conduction rate ratio
between superheated steam and saturated steam is probably equal
to the heat transfer coefcient ratio, that is 1/1501/250. Super-
heated steam carries more heat than saturated steam, and super-
heated steam heat loss rate is less than that of saturated steam
(Marx and Langenheim, 1959). It can take a relatively long time to
Table 2
Temperature and pressure in each region.
No. Region T P
( 1C) (MPa)
1 Cold water 0rTr350 PZP
s
(T)
2 Superheated steam 0rTr350 0oPrP
s
(T)
350rTr590 0oPrP
B
(T)
TZ590 P40
3 Wet steam 350rTrT
B
(P) PZP
B
(T)
4 Saturated steam 0rTr374 0oPr22.064
Fig. 1. Regions of water phase.
Wellhead thermal compensator
Superheated steam 480
o
C
tubing
Dry saturated steam
casing
water
Sealing Packer
Wet steam (dryness 50%
B
u
r
n
e
r

c
o
n
t
r
o
l

r
o
o
m
Superheater Unit
Heating to improve dryness
Boiler combustion chamber
S
t
e
a
m
-
w
a
t
e
r
s
e
p
a
r
a
t
o
r
s
90%)
B
u
r
n
e
r

c
o
n
t
r
o
l

r
o
o
m
Superheater Unit
Heating to improve dryness
Boiler combustion chamber
90%)
B
u
r
n
e
r

c
o
n
t
r
o
l

r
o
o
m
Superheater Unit
Heating to improve dryness
90%) Wet steam(dryness
)
Fig. 2. Superheated steam generated process.
A. Xu et al. / Journal of Petroleum Science and Engineering 111 (2013) 197207 199
cool, during which the steam is releasing very little energy and is
transmitted long distances, which is usefulfor increasing the
heating region. Superheated steam can overcome the limitation
that after 10 cycles of saturated steam stimulation, the maximum
heating radius is not enlarged.
Q AT T
w
1
Q
sup
Q
s

sup
AT
sup
T
w

s
AT
s
T
w

sup

s
T
s
T

T
w

T
s
T
w

sup

s
1
T

T
s
T
w


sup

s
2
3. Synthetic evaluating models for superheated steam owing
down the wellbore
Early mathematical model about calculation of steam para-
meters along the wellbore during steam injection were established
by Satter Ramey (Ramey, 1962), Satter (Satter Abdus,1965), and
Willhite respectively according to the conservation of energy and
heat conduction theory Willhite (1967), and this model does not
consider the variation of pressure during steam transmission in
the wellbore which leads to a greater error when the well depth is
deep or injection rate is high.
In the 1980s, Fontanilla and Aziz studied the impaction of the
steam pressure loss caused by gravity, frictionand acceleration and
established the integrated model by considering the steam ow
state, gas slippage of the two phase ow and heat transfer during
steam owing (Fontanilla and Aziz, 1982). In the1990s, Chiu
established the numerical model by an analytical method and
gave the following hypothesis: heat conduction from outer ring of
wellbore to cement is a one dimensional steady heat transfer
process, and heat from the outer ring cement to formation is a one
dimensional unsteady heat transfer process (Chiu and Thakur,
1991). Recently, some researchers used a more complex nite
difference method to solve the wellbore and the unsteady heat
transfer process from wellbores to the surrounding formation.
In this study, assumptions of the model foundation are as
follows: (1) Superheated steam ow is a constant mass ow in the
wellbore. (2) The bottom of insulated tubing is sealed with packer
setting to ensure no steam passes into annulus lled with air.
(3) The superheated steam owing in the wellbore is a one
dimensional steady ow. During superheated steam ow, pressure
and temperature in the same cross section are equal everywhere.
(4) Heat transfer from the inner surface of the tubing to the outer
ring of the cement is steady but from the outer ring of cement to
formation, the heat transfer is not steady, and the model is built
without considering heat transfer in the longitudinal direction
along the wellbore. The wellbore structure is shown in Fig. 4.
Table 3
The value of heat transfer coefcient.
Heat transfer conditions Value Commended value
(W/(m
2
K)) (W/(m
2
K))
Saturated steam heated or condensed 500015,000 10,000
Water boiling 100030,000 30005000
Water heated or cooling 2005000 4001000
Oil heated or cooling 501000 200500
Superheated steam heated or cooling 20100
Air heated or cooling 560 2030
4200
3800
4000
superheated steam
3200
3400
3600
2800
3000
quality x=1
2200
2400
2600
E
n
t
h
a
l
p
y
,

K
J
/
(
K
g
.
K
)
2000
5 5.5 6 6.5 7 7.5 8 8.5 9
saturated steam
Entropy, KJ/(Kg.K)
Fig. 3. H-S diagram of steam.
Fig.4. Wellbore structure.
A. Xu et al. / Journal of Petroleum Science and Engineering 111 (2013) 197207 200
The mathematical model is founded on the basis of three laws:
(1) the law of conservation of mass, (2) the theorem of momen-
tum, (3) the law of conservation of energy. Equations used in the
calculation are as follows:
1) Mass conservation equation
Superheated steam injection process is a constant mass ow
and the mass conservation equation is

1
v
1
A
2
v
2
A i
s
3
2) Theorem of momentum equation
dpA
m
Adzg cos
f

1
v
2
1
A
2
v
2
2
A 4
Eq. (4) describes that the superheated steam pressure drops along
the wellbore on the basis of analysis of innitesimal by taking into
account the gravity impulse, and
m
Adzg cos dt which is the
gravity impulse within the dt time. The new equation of pressure
drop dp was founded by the Theorem of Momentum.
3) Energy conservation equation
Considering frictional energy loss in the superheated steam
ow, the superheated steam energy equation was established
under the principle of conservation of energy:
dQ
dz

dW
dz
i
s
dh
m
dz
i
s
d
dz

2
2

i
s
g cos 5
The physical meaning of Eq. (5) is that the internal energy
change of microunits such as Fig. 5 in unit time as well as the
change of mechanical energy is equal to the heat transferred to
the wellbore and friction losses during the unit superheated
steam owing. Eq. (6) not only considers the mechanical and
internal energy changes of the microsuperheated steam unit
and the heat transferring of the micro-unit, but also considers
friction loss during the steam owing such as the energy dW,
which can be used to calculate the changes of superheated
steam temperature dT.
4) Auxiliary equation
dh
m
dz

h
m
T

p
dT
dz

dh
m
dp

T
dp
dz
6
The equation above can be translated into
dh
m
dz
C
p
dT
dz
V T
V
T

p
( )
dp
dz
7
Eq. (7) is applicable to all types of solid, liquid and gas.
5) Function relationship between saturated steam pressure and
temperature
t
s
210:2376p
0:21
s
30 8
Eq. 8 can determine whether the steam is superheated or not.
The mathematical model about calculation of superheated
steam parameters distribution along the wellbore was established
through Eq. (18).
3.1. Solution of mathematical model
The solution of this mathematical model was divided into four
parts: pressure drop calculation along the wellbore, radial heat loss
calculation, superheated steam temperature calculation, and super-
heated degree calculation. The solution procedure is shown in Fig. 6.
(1) The wellbore was divided into N nodes. The distance of each
segment was z. The wellhead was the rst node, i 1
Fig. 5. Mechanical analysis from a micro-unit. Fig.6. Solution procedure of equation variables.
A. Xu et al. / Journal of Petroleum Science and Engineering 111 (2013) 197207 201
(2)
In the rst node, if superheated degree T

40, for calculations

turn to next step 3. Otherwise, turn to step 6.
(3) From the beginning of node i, pressure and temperature drop
values along z were given as per the following equation,
p 0:1 z=100 MPa, T 5 z=100
1
C
(4) Step 3 was solved for p and T and substituted in Eqs. (4)(7),
to solve for dp and dT.
(5) If |dpp| o and |dTT| o were right, for calculations
turn to step 6, Otherwise, let pdp and TdT. Calculations
went back to step 4 until |dpp| o and |dTT| o
were right.
(6) If T

40 was right, calculations turned to step 7. Otherwise,

calculations were nished.
(7) Let i i 1 and calculations turn to next node, repeat steps
3 through 7 until the last node was completed.
In order to verify the accuracy of the model calculation results,
a steam injection well X16 was chosen to test the steam para-
meters (including pressure, temperature, and heat-loss) along the
wellbore. The parameters of the wellbore are shown in Table 4.
The superheated steam parameters on the wellhead such as the
steam injection pressure, steam temperature and superheated
degree were 3.4 MPa, 280.2 1C, and 39.1 1C respectively; and the
steam injection rate was 6 t/h. The comparison of test data and
calculation data curves are plotted in Figs. 710. The two types of
data curves had a good match and the deviation was less
than 3.12%.
4. Processes of heavy oil Reservoirs heated by superheated
steam
The changes of superheated steam can be divided into two
stages in heating the heavy oil reservoir. As is shown in Fig. 11, the
rst stage is superheated steam that has to be cooled to saturation
steam and the degree of superheat is to be reduced. When the
superheat degree reduces to 0, the superheated steam changes
into saturated steam, but the steam quality remains at 100%.
The second stage is when the steam temperature is kept constant
and the quality is decreased. In this stage, saturated steam begins
to condense to release its enthalpy of evaporation and its phase
Table 4
Parameters of wellbore structure.
Wellbore structure parameters
Values Wellbore structure parameters
Values
Inner tubing radius (m)
0.038 Outer tubing radius (m)
0.04445
thermal conductivity of heat insulation tube (W/(m 1C))
0.07 Surface roughness of tubing (m)
0.0000457
Inner casing radius (m)
0.0807 Temperature gradient of formation (1C/m)
0.029
Outer casing radius (m)
0.0889 Thermal conductivity of formation (W/m 1C))
1.73
Blackness of inner casing surface
1.0 Thermal conductivity of cement (W/(m 1C))
0.933
Outer cement radius (m)
0.213 Coefcient of temperature conductivity of formation (m
2
/h)
0.00037
Ground temperature (1C)
21
Fig.7. Comparison of steam temperature along the wellbore between tests and
calculations.
Fig.8. Comparison of superheated degree along the wellbore between tests and
calculations.
Fig.9. Comparison of steam pressure along the wellbore between tests and
calculations.
Fig.10. Comparison of heat loss rate along the wellbore between tests and
calculations.
Fig. 11. Changes of superheated steam in heating oil reservoir.
A. Xu et al. / Journal of Petroleum Science and Engineering 111 (2013) 197207 202
changes to liquid during the latent heat releasing and this causes
lowering of quality of saturated steam . When the steam quality
reduces to 0, the wet steam changes into hot water, but the steam
temperature remains the same as the dry saturated steam. As a
result, in the rst stage, from the well bore to the edge of
superheated steam heating area, the temperature gradually
decreases, and there exists a temperature gradient in the heating
region and on the heat transfer surface. In the second stage, due to
the release of latent heat and the uniform temperature of
saturated steam, temperature in the saturated steam heating area
is kept constant everywhere. In addition to all saturated steam
heating stages, superheated steam has its own superheated stage,
in which chemical reactions would happen such as formation of
water, heavy oil, and formation of minerals under the favorable
circumstances of high temperature of superheated steam. These
chemical reactions not only result in some changes of composition
and the irreversible reduction of its viscosity, but also change the
microscopic pore structure of rocks to improve the permeabilty of
superheated steam heating area; changing the wettability of the
reservoir rock, increasing the displacement efciency of super-
heated steam ooding, which bring about signicant improve-
ments of heavy oil development effects.
Fig. 11 also demonstrates an important dependence on the type
of steam, as the case with greater degree of superheat has greater
steam override. The greater the superheated degree is, the larger
the heated volume will be. Obviously, the scope of steam chamber
is controlled by steam override.
5. Mechanism of superheated steam huff and puff
5.1. Viscosity reduction under high temperature
As can be seen from the process of superheated steam heating
heavy oil reservoir, the overall heat transfer coefcient for super-
heated steam varies throughout the process, but will always be
much lower than that for saturated steam and its heat loss during
transmission is much less than saturated steam. There is a longer
time existing in the reservoir to heat heavy oil reservoir at a higher
temperature and a larger scope than saturated steam. The viscosity
of the heavy oil heated by superheated steam becomes lower than
that of heavy oil heated by saturated steam and the properties of
the heavy oil become as mobile as light oil. In certain pressure
gradients, the rates owing to the bottom hole were accelerated
and oil production was multiplied. Heavy oil pyrolysis depends on
the following three reaction equations:
Steam reforming:
C
x
H
y
2xH
2
O-xCO
2
(2xy/2)H
2
Water partial reforming:
COH
2
O-CO
2
H
2
Heavy oil desulfurization:
C
x
H
y
S
2
H
2
O/H
2
-2H
2
SLight hydrocarbon mixture
Dielectric constant of superheated steam is almost the same as
in a weak polar solvent and the capacity of dissolving organic from
heavy oil is very strong. Light hydrocarbon can be produced after
the above three reactions . If the temperature was higher , the
reactions could occur easily. The following tests were important
evidences to describe the heavy oil composition change. The three
heavy oil samples were from three different reservoirs in Post-salt
Oil Field marked by the symbols of 1#, 2#, 3#. The heavy oil
composition was analyzed by gas chromatography and the data of
hydrocarbon distribution was processed by the normalization
method. Table 5 summarizes the composition changes of the three
samples after chemical reactions under the temperature of 300 1C
lasting 70 h. The saturated hydrocarbon content was increased
from 27.24%, 25.61% and 30.52% to 30.91%, 28.68% and 33.31%
respectively. The aromatic hydrocarbon content also increased by
3.81% on average. Compared with the original oil, the content of
resin decreased by 3.97%, 8.59% and 3.40%. Also the content of
asphalt was reduced by 1.74%, 2.01% and 1.24%. Therefore, light-
weight components were signicantly increased and heavy com-
ponents obviously reduced after heavy oil was treated by
superheated steam. These amounts of lightweight components
could make the viscosity of heavy oil decrease fundamentally.
5.2. Reservoir properties and pore structure reformed by
superheated steam ooding
Unlike saturated steam, the temperature of superheated steam
is not uniform. The temperature for superheated steam varies
everywhere throughout the heating process. This means that
superheated steam creates temperature gradients on the heat
transfer surface as it cools to saturation temperature (Marx and
Langenheim, 1959). The high temperature gradient that would
have a strong role to damage the reservoir rock and some clay
minerals are migrated to the well bottom, some in micro-particle
size are dissolved by superheated steam ooding under high
temperature and high pressure conditions. As a result, the original
pore structure was reformed and micro-pore throats were
destroyed by superheated steam ooding. The resistance of the
owing reservoir uid decreased signicantly, which made reser-
voir permeability increase greatly. According to the X ray diffrac-
tion analysis of 46 core samples, the distribution of clay minerals
Table 5
Comparison of the three heavy oil samples composition changes.
Sample
1# 2# 3#
Original oil 300 1C 72 h treatment Original oil 300 1C 72 h treatment Original 300 1C 72 h treatment
SARA, W/%
Saturated 27.24 30.91 25.61 28.68 30.52 33.31
Aromatic 31.55 33.59 25.75 33.28 36.41 38.26
Resin 34.36 30.39 41.92 33.33 28.51 25.11
Asphalt 6.85 5.11 6.72 4.71 4.56 3.32
Hydrocarbon distribution of the samples W/%
oC10 0.11 5.38 0.24 6.06 3.3 7.27
oC15 6.32 23.28 7 24.35 14.19 27.74
oC20 13.28 38.83 15.38 37.17 20.9 53.96
oC25 17.91 51.01 21.7 46.27 28.32 55.97
oC30 56.74 76.06 65.37 77.29 73.4 83.27
oC35 93.91 98.55 97.31 99.07 98.57 99.58
A. Xu et al. / Journal of Petroleum Science and Engineering 111 (2013) 197207 203
such as kaolinite and illite montmorillonite mixed are stable at the
vertical and horizontal direction in the Post-salt reservoirs in
Kenkiyak. For the original oil sands, the surface was uneven and
adhered with many micro-crystals. After supperheated steam
ooding, the surface of oil sands become smooth, as shown in
Fig. 12 and log interpretation results showed that the porosity
increased by 12%, and permeability increased to 300400%, as
shown in Fig. 13.
5.3. Change of wettability of rock
The wettability tests at high temperature indicated that with
temperature increasing, oilwater interfacial tension was reduced,
wettability of the rock shifted from oleophilic to hydrophilic or
else from weak to strong hydrophilic (Martin and Jone,1967).
The mechanism of this effect is that polar materials such as resin
and asphaltene are very rich in heavy oil and the polar molecules
adsorbed on the oilwater interface or the rock surface in low
temperature; therefore, the majority of cores are oleophilic. When
the temperature increases, polar material is discharged gradually
and cores become hydrophilic, and oil permeability is increased,
which will help improving the oil displacement efciency.
Three different types of wettability medium such as glass
beads, oil sand, quartz sand were ooded by superheated steam
to test the oil displacement efciency. The results showed that the
stronger the water was wet, the higher the displacement efciency
increased. Oil displacement efciency by superheated steam drive
was 612% more than that of ordinary saturated steam thermal
recovery, which is shownin Table 6.
6. Superiority of superheated steam huff and puff
The Kenkiyak post-salt oil eld is a heavy oil reservoir with the
average porosity of 36.6%, average permeability around 187510
3
m
2
, buried depth between 290 and 380 m, dead oil viscosity
within 144691 mPa s at 20 1C and the reservoir temperatures
between 18.8 and 20 1C. The Pre-salt oil eld is situated on top of
the salt dome of the Kenkiyak district. The target formation is the
Middle Jurassic formation. The reservoir is mainly made up of
medium to ne sandstones, and silt ne sandstones with 6.54% clay
minerals. The reservoir rocks are unconsolidated, being mainly
porous cement. The Middle Jurassic formation is divided into two
sub groups, J-I and J-II, which have the same oil/water contact with
the depth of 380 m. The oil/water relationship is simple with the
upper part of the structure being the pure oil zone and the lower
part the water zone. The oil layers are well separated, providing
favorable conditions for separate layer development.
The burial depth of the Middle Jurassic formation ranges from
290 to 380 m. The oil formation is thick with the average
effective thickness of 17.03 m and the average net/gross thickness
ratio is over 0.70.
The Middle Jurassic formation is a typical shallow, medium to
massive laminated structural lithologic conventional heavy oil
reservoir with edge water high porosity and high permeability.
The average porosity and permeability are 32.5% and 1875 10
3
m
2
respectively.
Fig. 12. Changes of reservoir rock after.
Fig. 13. Permeability changes after superheated steam ooding.
Table 6
Oil displacement efciency.
Medium Saturated steam Superheated steam 1 Superheated steam 2
(%) (%) (%)
Glass beads 68.2 69.1 69.1
Oil sands 76.5 85.3 85.8
Quartz 71.7 84.3 86.3
Note: Superheated steam 1 (superheat 39 1C) and superheated steam 2 (superheat
70 1C).
A. Xu et al. / Journal of Petroleum Science and Engineering 111 (2013) 197207 204
Table 7 shows the reservoir parameters and uid properties of
the Middle Jurassic formation in Pre-salt oil reservoir.
The oil properties are as follows: (l) the oil density is 0.987 g/cm,
at 20 1C, (2) the resin and asphaltene content is about 33.3%, (3) the
wax content is 2.2%. According to the heavy-oil classication system
in China, it is ordinary heavy oil.
Pre-salt oil eld has been produced using cyclic saturated
steam injection since 1983 and been shifted to saturated steam
ooding in main part of reservoirs between 1984 and 1996. The
original well spacing has reduced from 14l m to 100 m between
1984 and 1997. In some areas, well spacing is now 70 m. At the end
of 2009, the cumulative oil recovery was 14.3% of the OOIP.
A three dimensional simulation model was built using CMG
STARS and was tuned with experimental data from well 43. The
model consisted of a vertical matrix block divided into 24 grids in
Z direction, 15 grids block in X direction, and 15 grids block in Y
direction. Total matrix block distance was 13.3 m with 0.808 m of
width and depth in the X and Z directions (Kuo and Phocas, 1969).
The model is homogenous and the parameters are summarized in
Table 8. Dead-oil viscosity versus temperature and the relative
permeability curves follow in Table 9 and Fig. 14, respectively.
History matching of oil production from a single well was rst
conducted according to the actual production of well 43. After a
production history matching, model recovery factor was 14.5% and
the average remain oil saturation was 67.7%. Sensitivity of the
results to input values of the temperature and superheated degree
of superheated steam, the scope of formation heated by super-
heated steam were studied and parameters such as cumulative oil
produced, amount of oil produced were also investigated.
The saturated steam huff and puff process was also investigated
and compared with the results of superheated steam huff and puff
on the basis of the same single-well model. The well 43 went
through in sequence all the three production stages such as by
natural depletion, by saturated steam huff and puff, and by
superheated steam huff and puff. The injection of superheated
steam and saturated steam were under the same pressure but the
temperatures were different, what were chosen so as to resemble
the average eld injection conditions. For the rst cycle 2600 t of
steam was injected. The superheated steam injection rate was
150 t/d.
In order to analyze the effects of different steam huff and puff,
several different types of steam ,which carried the same quantity
of heat and had different temperatures, qualities and degrees of
superheat, were injected into the same oil reservoir. For the rst
cycle, the quantity of steam injected and cumulative oil production
for different steam huff and puff are shown in Table 10. Super-
heated steam huff and puff results in signicantly greater produc-
tion. After roughly 700 days of production, superheated steam huff
and puff produced about 150% more oil than saturated steam. In
the superheated steam huff and puff case, the cumulative oil
production is 4050 t and the OSR (oil steam ratio) is 1.8 at 700 days.
The cumulative oil production is 1463 t greater than that of saturated
steam due to extra oil production associated with the injection of
superheated steam, which reected the superiority of superheated
steam huff and puff.
Fig. 15 demonstrates an important dependence on the type of
steam in that the case with the greater degree of superheat has a
greater steam override. The heated volume is larger at greater
degree of superheat. Obviously, the scope of steam chamber is
controlled by steam override. The heat radius of superheated
steam in the rst cycle reached 30 m, about 10 m larger than that
of saturated steam.
7. Production performances of superheated steam huff and
puff application
From 1967 to 2002, the Post-salt oil reservoir was developed by
depletion. As a result, the average reservoir pressure has dropped
from 5.9 to 1.8 MPa. The serious heterogeneous characteristics and
Table 7
Reservoir parameters and uid properties of the Middle Jurassic formation.
Average effective
thickness
OOIP Porosity Perm. Original oil
saturation
Dead oil viscosity
mPa s at 50 1C
Oil density
(20 1C)
Formation
temperature
Original formation
pressure
(m) (10
4
t) (%) (10
3
m
2
) (%) (g/cm
3
) (1C) (MPa)
17 54 2138 2.628840 65 144691 0.9053 18.8 2.82
Table8
Parameters for the single-well model.
Reservoir depth 280 m
Initial reservoir pressure 2.82 MPa
Net pay 17 m
Porosity 36.60%
Permeability 1875 md
Oil saturation 65%
Dead oil viscosity 269 cp
Table 9
Dead-oil viscosity versus temperature.
Temperature (1C) 20 30 50 80 150 200
250 280 300
Oil viscosity (mPa. s) 262 134 44 14 3.6 2.2
1.6 1.2 0.9
0
0.2
0.4
0.6
0.8
1
0 0.2 0.4 0.6 0.8
Sw
K
r
o
0
0.2
0.4
0.6
0.8
1
K
r
w
Kro
Krw
Fig.14. Relative permeability of oil and water.
Table 10
Different steam carried the same heat huff and puff effects.
Temperature Quality Degree of
superheat
Duration Cyclic
injection
Cyclic
production
OSR
(1C) (%) (1C) (d) (t) (t) (fraction)
236 40 0 700 3716 1713.7 0.46
236 60 0 700 3114 1720.1 0.55
236 80 0 700 2680 1980.6 0.74
236 100 0 700 2352 2587.2 1.10
286 100 50 700 2250 4050.0 1.80
A. Xu et al. / Journal of Petroleum Science and Engineering 111 (2013) 197207 205
rapid water invasion made the water-cut of wells as high as 76%.
Furthermore, the average well production rate declined from the
original 8 m
3
/d to 12 m
3
/d, and has stayed at this level for a long
time. The eld was hovering at the economical margin in 2002.
In 2003, the operator launched a series of studies and pilot tests
to improve the production performance and economic benets.
Pilot tests were performed covering the saturated steam huff
and puff in 28 wells and along with the superheated steam huff
and puff there were in total 80 wells . The superheated steam is
more effective than the saturated one in heating water-invaded oil
reservoirs. Among all the pilot test wells, 13 went through in
sequence all the three production stages, i.e. by natural energy, by
saturated steam huff and puff, and by superheated steam huff
and puff. First by natural energy, the daily production was only
12 m
3
/d, which is far from the economic limit. Later by saturated
steam huff and puff, the average daily production increased to
3 m
3
/d, which was still not ideal. Two years later by superheated
steam huff and puff, the mean daily production was raised to
89 m
3
/d, which increased the daily oil production greatly. The
comparison is shown in Fig. 15. The production period with
superheated steam huff and puff has lasted over 700 days in the
rst cycle, and is still extending ahead. With high initial daily oil
production by superheated steam huff and puff, the average increase
of oil was 6.4 t/d, which was 2.8 t/d higher than the saturated steam.
Cyclical oil production reached 4160 t by superheated steam huff and
puff, which was 2192 t more than that of saturated steam stimula-
tion. Cyclical oilsteam ratio was 1.8 by superheated steam huff and
puff, 0.7 higher than that of saturated steam.
When converted to superheated steam huff and puff, the
formation temperature kept rising and the middle and upper
parts of the formation were heated with the maximum tempera-
ture reaching 220 1C in the inter-well formation, which reected
the superiority of superheated steam huff and puff.
Horizontally, the formation with the temperature greater than
150 1C covers a distance of 1020 m and the steam swept radius is
about 30 m, 10 m greater than that of saturated steam due to extra
oil production associated with the injection of superheated steam,
as is shown in Fig. 16.
8. Results and conclusions
Superheated steam exists at a temperature higher than that of
saturated steam without the limitation of pressure, has a higher
temperature, carries more heat and has greater heating capacity
than saturated steam. Superheated steam is always in the state of
highest quality, 100%, which determines that it has a very small
heat transfer coefcient. The heatloss of superheated steam during
transmision in wellbores is 1/1501/250 as much as that of
saturated steam,which means that much more heat is carried to
the heat oil reservoir and at the same time it reaches further
distance for superheated steam. Under the condition of carrying
the same heat, heating radius by superheated steam huff and puff
is about 10 m longer than saturated steam. Superheated steam
huff and puff was put into Kazakstan's heavy oil reservoir after
saturated steam huff and puff and the average daily oil production
was 24 times that of saturated steam huff and puff, which thus
improved heavy oil production effectively. Superheated steam huff
and puff as a secondary thermal recovery is very appropriate for
difcultly developed heavy oil reservoir.
References
Chang, Yuwen, Zhang, Yi, Hu, Yongjiu, et al., 1997. The Progress of Heavy Oil
Thermal Recovery Technology [M]. Petroleum Industry Press, Beijing.
Chen, Minfeng, Lang, Zhaoxin, Mo, Xiaoguo, 2002. Optimization of steam stimula-
tion injection parameter and the reasonable development limits of ultra-heavy
oil. J. Univ. Pet. China (Edition of Nat. Sci.) 26 (1), 3942.
Chiu, K., Thakur, S.C., 1991. Modeling of wellbore heat losses in directional wells
under changing injection conditions[C]. SPE 22870.
Fontanilla, J.P., Aziz, K., 1982. Prediction of bottom-hole conditions for wet steam
injection wells. JCPT, 8288.
Kuo, C.H., Phocas, D.M., 1969. A gravity drainage model for the steam-soak process
[C]. SPE2329, pp. 119126.
Li, Chuntao, Qian, Genbao, Wu, Shuhong, et al., 2008. The property of superheated
steam and its application in the development of heavy oil reservoirs: Kenkiyak
Oil Field in Kazakhstan as an example. Xinjiang Pet. Geol. 29 (4), 495497.
Marx, J.W., Langenheim, R.H., 1959. Reservoir heating by hot uid injection. Trans.
AIME 216, 312315.
Martin, Jone, C. 1967. A Theoretical Analysis of Steam Stimulation, JPT, March 4,
pp. 411418.
Qu, Tongci, Li, Shuang, 2009. Technique of enhancing recovery of remaining oil in
heavy oil reservoirs by horizontal wells. Pet. Explor. Dev. 36(6), 743748.
Rohsenow, Warren M., et al., 1992. Application of Heat Transfer Handbook. Science
Press, Beijing, pp. 102135.
Ramey, H.J., 1962. Wellbore heat transmission. JPT 4. (427435).
Satter Abdus, 1965. Heat losses during ow of steam down a wellbore. JPT (7), 845851.
Shen, Weidao, Jiang, Zhimin, Tong, Jungeng, 2000. Engineering Thermodynamics.
Higher Education Press, Beijing, pp. 212229.
Fig. 15. Distribution of temperature after the same heat of different steam injection in one cycle. (a) Superheated steam huff and puff and (b) saturated steam huff and puff.
0
3
6
9
12
15
46 43 54 39 41 16 56 33 49 45 35 15 19
well
o
i
l

r
a
t
e
(
t
/
d
)
production on natural energy
saturated steam huff and puff
superheated steam huff and puff
Fig. 16. Oil rate of single well under three different recovery methods.
A. Xu et al. / Journal of Petroleum Science and Engineering 111 (2013) 197207 206
Wu, Xianghong, Xu, Anzhu, Fan, Hailiang, 2010. An integrated evaluation on factors
affecting the performance of superheated steam huff and puff in heavy oil
reservoirs. Pet. Explor. Dev. 37 (5), 608613.
Willhite, G.P., 1967. Overall heat transfer coefcients in steam and hot water
injection wells JPT, 607617.
Yan, Ke, Ren, Huaiqiang, 2009. Oil/water inversion and juxtaposition in heavy oil
reservoirs: taking 67 sand members of Guantao Formation, Gudao Oileld as
an example. Pet. Explor. Dev. 36 (5), 635640.
Zhang, Yitang, Li, Xiuluan, Zhang, Xia, et al., 2008. Four fundamental principles for
design and follow-up of steam ooding in heavy oil reservoirs. Pet. Explor. Dev.
35 (6), 715719.
Further reading
Anon, 2008. Pet. Explor. Dev. 35 (1), 113117.
Boberg, T.C., Lantzs, R.B., 1966. Calculation of the production rate of a thermally
stimulated well. J. Pet. Technol., 16131623.
Farouq, S.M.A, 1981. Comprehensive well bore steam/water ow model for steam
injection and geothermal applications. SPEJ, 527534.
The International Association for the Properties of Water and Steam. IAPWS-IF97,
1997. International Formulation 1997 for the Thermodynamic Properties of
Water and Steam [R]. Erlangen, Germany.
A. Xu et al. / Journal of Petroleum Science and Engineering 111 (2013) 197207 207