PETROLEUM EXPLORATION AND DEVELOPMENT

Volume 44, Issue 6, December 2017

Online English edition of the Chinese language journal

Cite this article as: PETROL. EXPLOR. DEVELOP., 2017, 44(6): 1016–1021. RESEARCH PAPER

Gravel sizing method for sand control packing in hydrate

production test wells
LI Yanlong1, 2, HU Gaowei1, 2, *, LIU Changling1, 2, WU Nengyou1, 2, CHEN Qiang1, 2, LIU Lele1, 2,
LI Chengfeng1, 2
1. Key Laboratory of Gas Hydrate, Ministry of Land and Resources, Qingdao Institute of Marine Geology, Qingdao 266071, China;
2. Laboratory for Marine Mineral Resources, Qingdao National Laboratory for Marine Science and Technology, Qingdao 266071, China

Abstract: To deal with sand production problems during the process of producing natural gas from hydrate-bearing sediments (HBS)
using reservoir-fluid extraction method, a new gravel sizing method for sand control packing named “Hold coarse while eliminate fine
particle (HC & EF method)” was developed for the clayey hydrate-bearing formations. Site X, in Shenhu area, South China Sea was
taken as an example to describe detailed gravel sizing procedure. On the basis of analyzing basic particle size distribution (PSD) charac-
teristics of HBS at Site X, the formation sand was divided into two components, which are coarse component and fine component. The
gravel sizes for retaining coarse component and eliminate fine component were calculated, respectively. Finally, intersection of these two
gravel sizes was taken as the proper gravel size for Site X. The research results show that the formation at Site X is clayey sand with poor
sorting and uniformity, proper gravel size for upper segment packing is 143215 μm, while that for lower segment packing is 240360
μm. In consideration of the difficulty of layered sand control operation on offshore platform, proper gravel packing size for Site X is
recommended as 215360 μm.

Key words: gas hydrate; production test; sand management; gravel sizing; South China Sea; Shenhu area

Introduction With shallow burial depth, weak consolidation and high

Previous studies have proved the existence of huge amount
of natural gas hydrate in Northern South China Sea[12], Chi-
na’s first hydrate production led by China Geological Survey
(CGS) has been developed in 2017 in Shenhu area. After the
successful ignition on May 10th by Ministry of Land and Re-
sources, around 12×104 m3 of natural gas in total was
extracted from hydrate-bearing sediment (HBS), with
methane fraction up to 99.5% and average daily production of
1.6×104m3 till May 18th, which make China the first country
achieve continuous and stable natural gas hydrate production
from marine silt sand reservoirs[3]. The production continued
until July 9th, with cumulative gas production of 30.9×104 m3
and average daily production of more than 5 000 m3[4]. Huge
amount of production test data was obtained for further study.
Sand production is unavoidable during hydrate production
because that hydrate reservoirs are usually unconsolidated,
poorly consolidated or fracture developed formations[56],
especially for marine ultra-fine silt sand HBS in South China
Sea [78]. Sand management strategy plays an important role in
extending efficient hydrate production test.
shale content, HBS in Northern South China Sea is uncon-
solidated ultra-fine silt reservoirs[9], gravel packing (including
open hole gravel pack, internal gravel pack and ceram-
site/quartz pre-packed screen) is one of the most efficient
choices for this kind of reservoir[1011]. Under gravel packing
condition, the packing layer acts as both sand blocking barrier
and flow channel of reservoir fluid. Larger gravel size is good
to lower additional skin factor and improve well productivity,
but could lead to vast sand production, sand burial and forma-
tion depletion at the same time. On the other hand, small
gravel size is good for sand retention but may lead to block-
age by fine or shale particles, which is fatal for well produc-
tivity. Therefore, the design of gravel size for gravel packing
should take into account both the requirement of sand control
and avoiding blockage sand retention layer.
At present, the design of gravel size for packing often fol-
lows two ideas: one is to retain formation sand entirely, in
which the solid content of produced fluid is less than 0.3‰;
the other is moderate sand-control, in which solid content of
produced fluid at wellhead is less than 0.5‰[12]. However,

Received date: 20 Jun. 2017; Revised date: 31 Aug. 2017.

* Corresponding author. E-mail: hgw-623@163.com
Foundation item: Supported by National Natural Science Foundation of China (41606078); National Key Research and Development Plan (2017YFC0307600); and
Qingdao National Laboratory for Marine Science and Technology (QNLM2016ORP0203, QNLM2016ORP0207).
Copyright © 2017, Research Institute of Petroleum Exploration and Development, PetroChina. Published by Elsevier BV. All rights reserved.
 LI Yanlong et al. / Petroleum Exploration and Development, 2017, 44(6): 1016–1021
shale content of clayey silt HBS is more than 30%, shale and
fine particles are the main content that may lead to blockage
of sand control packing[1314]. The solid content in wellhead
production fluid is supposed to be larger than that in conven-
tional oil and gas well in order to guarantee the plugs can be
expelled from sand retention layer. Consequently, both en-
tirely sand-control and moderate sand-control design ideas are
not suitable for marine clayey silt HBS.
In this paper, a new gravel sizing method is proposed spe-
cially for hydrate production test wells in marine clayey silt
HBS. A typical marine site X in Shenhu area, Northern South
China Sea is taken as an example to illustrate the detailed
design procedure of the above gravel sizing method. Finally,
suitable gravel size for site X is suggested.
1. Gravel sizing method
In light of the specific characteristics of marine clayey silt
HBS, the main objectives of gravel sizing for hydrate exploi-
tation test well is: to ensure discharge of formation plugs, and
prevent coarser particles entering wellbore from formation.
Therefore, a new gravel sizing method based on the above
criterion called “hold the coarse while discharge the fine
(HC&DF method)” is advanced. In other words, by discharg-
ing fine particles to prevent flow channel blockage and stop-
ping coarse particles out of the packing layer, HC&DF
method can clear the near wellbore formation and ensure gas
production, and prevent wellbore collapse due to massive
sand production.
The procedure of HC&DF gravel size design is as follows.
Firstly, analyze the uniformity and sorting characteristics of
the HBS via original particle size distribution (PSD) parame-
ters, and set preliminary requirements for gravel sizing. Sec-
ondly, the formation particles are divided into two groups as
coarse content and fine content by mathematical conversion of
the PSD curve, and the cutoff point of these two groups is
worked out, which is also the maximum size of fine particles.
This value can be used to calculate the minimum gravel size
which can ensure discharge of fine sand content. Thirdly, dis-
lodge the fine segment from original PSD curve, new PSD
curve only for coarse component can be obtained, and the
distribution characteristics of coarse component are analyzed.
After that, the gravel sizing model based on entirely
sand-control theory is used to calculate the range of gravel
size to prevent the coarse component entirely. Finally, we get
two sets of size ranges determined by fine component dis-
charge and coarse component retention. Intersection of these
two sets is taken as the proper gravel size for corresponding
layers.
It is noteworthy that the above described HC&DF method
should also take the other factors, such as shale content and
layered sand-control into consideration. The application of the
obtained gravel size should be matched with the industrial
gravel size.
 1017 
2. Original PSD characteristics
PSD characteristics of produced layer are the base for
proper gravel size design of packing. The HBS in Northern
South China Sea has the features of shallow depth, low per-
meability, weak consolidation and high shale content. Original
PSD range of HBS in Site X is shown in Fig. 1, it is obvious
that the median grain diameters fall between 6.0 μm and 15.9
μm. The shale content of HBS for Site X is around 25%36%,
in which montmorillonite fraction is about 38% and illite frac-
tion is around 32%, so HBS at site X can be classified into
clayey silt[15]. Moreover, engineering geological data show the
particle size of HBS at Site X increases with the increase of
burial depth. The left boundary of the typical PSD curve in
Fig. 1 represents the typical grain size of the lower section of
the reservoir, and right boundary of the typical PSD curve
represents the typical grain size of the upper section of the
reservoir. The overall grain size distribution of Site X falls in
between these two typical curves.
Except median grain diameter, proper gravel sizing design
should also consider the influence of sorting coefficient and
uniformity coefficient[16]. Sedimentology’s formula Eq. (1)
and Berg’s correlation Eq. (2)[17] are commonly used to calcu-
late sorting coefficient:
84  16 95  5
F  (1)
4 6.6
where, 84  log 2 d84 , 16  log 2 d16 , 95  log 2 d95 ,
5  log 2 d5 .
90  10
F (2)
2
where, 90  log 2 d90 , 10  log 2 d10
According to Eqs. (1)-(2), the sorting standard of formation
sand can be divided into four levels: well sorted (F0.5),
normal sorted (0.5<F1.0), poor sorted (1.0<F2.0) and ex-
tremely poor sorted (F>2.0)[17]. Then, sorting coefficients of
HBS at Site X calculated by Sedimentologist’s formula and
Berg’s correlation are 1.852.30 and 2.352.95 respectively.
On the whole, the sorting coefficient obtained from Berg’s
correlation is higher than that obtained from Sedimentolo-
gist’s formula. But both of them indicate that sands of HBS at
Fig. 1. Original PSD curves of HBS at Site X.
 LI Yanlong et al. / Petroleum Exploration and Development, 2017, 44(6): 1016–1021
Site X are poorly sorted, which increases the difficulty of sand
retention operation. Because the poorer the sorting of the for-
mation sand, the larger the gravel size range needed for sand
retention will be, which lead to poor sand prevention effi-
ciency and anti-blockage of the gravel layer.
Uniformity coefficient of formation particles can be calcu-
lated by Eq. (3)[18]:
d 40
C
d90
Uniformity of formation sand can be divided into three lev-
els according to Eq. (3): uniform sand (C5), non-uniform
sand (5<C8) and extremely non-uniform sand (C>8). Uni-
formity coefficient of HBS at Site X is 8.412.3, suggesting
HBS belongs to extremely non-uniform sand.
Overall, HBS at Site X can be defined as extremely
non-uniform clayey silt with high shale content and extremely
poor sorting, which pose great challenges to sand-control.
3. Properties analysis of coarse component and
fine component
3.1. Division of coarse component and fine component
When examining the invasion of formation sand to well-
bore, Markestad et al.[19] found that a sand sample could al-
ways be divided into two components, the fine and coarse
contents. For the clayey silt HBS in Northern South China Sea,
when hydrate decomposes completely, the fine components
and clay minerals can be regarded as existing mainly in free
state in pores formed by coarse component. Driven by the
pressure differential, fine components carried by water/gas
would flow toward the packing layer. In order to balance the
productivity and formation stability, on the one hand, the fine
components must be discharged away to lower near wellbore
pollution degree, facilitate hydrate decomposition and en-
hance productivity by fluid extraction method; on the other
hand, production of coarse component must be prevented to
maintain the basic formation skeleton and avoid large area
formation voidage.
The basic method for formation sand partition is as follows:
(1) In order to get the critical size for partition of coarse and
fine contents, particle size analysis should be based on particle
number instead of mass fraction. Mass fraction is transferred
Fig. 2. Particle size partition of coarse component and fine component of HBS at Site X.
into particle number according to corresponding particle size
and density with the base of PSD curve as Eq. (4)[18].
MW
N (4)
4 
  πd 3 
3 
(2) A new PSD curve that demonstrates relationship be-
tween particle diameter and log(N) can be derived, as shown
in Fig. 2. The typical d-lgN relationship of upper section and
(3)
lower section of reservoir in site X is as Fig. 2.
(3) It can be seen from d-lgN curve that lgN decreases with
the increase of d and there existing a critical size, that is, when
particle size is lower than the critical size, lgN decreases rap-
idly with the increase of particle size. However, when particle
size is larger than the critical size, lgN decreases slowly with
the increase of particle size. In this paper, the above men-
tioned critical particle size is defined as the cutoff point be-
tween coarse component and fine component.
It can be seen from Fig. 2 that cutoff points of upper section
and lower section are 5.3 μm and 7.2 μm, respectively.
3.2. Characteristics analysis of coarse component
After partition of fine and coarse components, PSD law
analysis for formation sand dislodged fine content needs to be
taken, some critical characteristics should be worked out for
the coarse component to provide basis for gravel size design.
According to the partition results of coarse and fine com-
ponents shown in Fig. 2, the fine components with sizes less
than 5.3 m and 7.2 m respectively are removed from upper
and lower sections of reservoir in Fig. 1. Then new PSD
curves can be obtained for only the coarse component at Site
X, as shown in Fig. 3.
Based on Fig. 3, the coarse component of HBS at Site X
has a median grain size range between 15.9 μm and 24.4 μm,
sorting coefficients calculated by Eq. (1) and Eq. (2) of
0.91.2 and 1.11.5, respectively, and uniformity coefficient
of 1.982.98. It can be seen from the above parameters that
both the sorting coefficient and uniformity coefficient de-
crease after the fine component is removed from the original
formation, which means the formation with better sorting and
uniform sand. Thus the difficulty of sand control is lowered,
and the sand control design idea of discharging fine compo-
 1018 
 LI Yanlong et al. / Petroleum Exploration and Development, 2017, 44(6): 1016–1021

ponent, gravel size for blocking coarse component can be

calculated by the normal gravel sizing models.
Gravel sizing models commonly used include Karpoff
method, Tausch & Corley method, Saucier method and so on.
Saucier method and Tausch & Corley method use d50 and d10
on PSD curve respectively, to calculate proper gravel size
range for the packed layer. The calculation equations are Eqs.
(9) and (10) respectively:
 Dcmin  5d 50
 (9)
 Dcmax  6d50
 Dcmin  4d10
Fig. 3. PSD curves of coarse component of HBS at Site X.  (10)
 Dcmax  6d10
nent is conducive to prolonging term of sand control validity
Karpoff model determines gravel size range based on
and production test cycle.
uniformity coefficiend and d50 .
4. Gravel sizing of packed layers  Dcmin  5d50 , Dcmax  10d 50 C  3
 (11)
4.1. Minimum gravel size for discharge of fine component  Dcmin  4d50 , Dcmax  8d50 C  3
Based on Fig. 3 and Eqs. (9)-(11), proper gravel sizes for
After the study on relationship between gravel-sand ratio
blocking coarse component can be predicted, which are listed
(GSR) and gravel layers permeabilities through systematic
in Table 1.
drainage experiment, Saucier[20] found that, irrespective of
shale content, formation sand can pass through the gravel 4.3. Optimum gravel size for sand control at site X
layer smoothly when GSR is greater than 14. Therefore, the
For hydrate exploitation wells, proper gravel size should
minimum gravel size for the packed layer in hydrate exploita-
meet the requirements of discharging fine component and
tion wells should be:
retaining coarse component simultaneously. If the gravel sizes
  14d fmax
Dfmin (5)
for retaining coarse component and discharging fine compo-
Moreover, the minimum gravel size for upper section and
nent are viewed as two independent datasets, intersection of
lower section of site X can be calculated by Eqs. (6) and (7),
the two is the proper gravel size for gravel packing. Proper
respectively.
gravel sizes for site X are shown in Table 2. Empty sets in
  14  5.3 μm  74.2 μm
Dfmin (6)
Table 2 indicate the gravel sizing method is not suitable for
  14  7.2 μm  100.8 μm
Dfmin (7) clayey silt hydrate exploitation wells while the intersection
Besides fine component, there existing large amount of clay value is the optimum gravel size.
minerals like montmorillonite in clayey silt hydrate bearing It can be seen from Table 2 that Saucier method is not suit-
formation, which is the main factor causing blockage of the able for clayey silt hydrate reservoir. The gravel sizing result
packed layer[21]. The higher the shale content, the more serious from Karpoff method is conservative and that from Tausch &
and the faster the blockage of packed layer will be. Ma Shu- Corley method is a little bit risky. Decision of final packed
ai[22] found that with the rise of shale content, the optimum gravel size should be based on balance of productivity and
gravel size needed for controlling sand of the same median sand control effect. On one hand, if wellbore sand-carrying
grain size and uniform coefficient increases. When the shale condition is rigorous, productivity should be sacrificed partly,
content reaches 21.36%, the optimum gravel size for entirely
sand-control is 1.54 times of that when the shale content is Table 1. Gravel size for blocking coarse component at site X.
zero. Therefore, the concept of loss factor of sand retention Gravel size/μm
precision in gravel pack(denoted as Rm) is introduced to de- Section Tausch & Corley
scribe the influence of shale content. Increase of shale content Saucier method Karpoff method
method
will lead to larger size of gravel packing paricles. The maxi-
Upper 143215 6982 69137
mum gravel size with consideration of shale content for
Lower 240360 122146 122244
packed layer can be written as Eq. (8).

Dfmin  Rm Dfmin (8)
Table 2. Gravel sizing result of site X.
The minimum gravel sizes for discharging fine component
Gravel size/μm
of the upper section and lower section at site X are 118.7 μm
Section Tausch & Cor- Saucier
and 161.3 μm, respectively. Karpoff method
ley method method
4.2. Design of gravel size for holding coarse component Upper [143, 215]  [118.7, 137.0]
Under prerequisite of discharging all possible fine com- Lower [240, 360]  [161.3, 244.0]

 1019 
 LI Yanlong et al. / Petroleum Exploration and Development, 2017, 44(6): 1016–1021
then gravel sizing result from Karpoff method should be taken.
On the other hand, the result from Tausch & Corley method
will be suggested without rigorous sand-carrying condition
and with liquid flowing into pipeline. It is worth noting that
all the above research results are based on the assumption that
the gravel packed layer can retain all coarse component.
Therefore, if sand bridge can form stably at the bottome hole,
the result from Tausch & Corley method is enough to retain
coarse component theoretically. Therefore, the result based on
Tausch & Corley method was recommended here for calyey
silt hydrate exploitation wells in South China Sea. The
suggested gravel packing size for upper section at site X is
143215 μm, while that for lower section is 240-360 μm. But
on offshore platform, it is difficult to implement layered grav-
el packing because of complex operation procedure, so the
above sand packing design results can’t be followed strictly.
In order to reconcile the upper section and lower section, the
gravel size range is suggested at 215-360 μm with the condi-
tion of meeting filling strength. The design value could be
matched with industrial gravel size during application.
It should be noted that the gravel sizing method provided in
this paper is properly used for inner-hole gravel pack
operation or pre-packed screen operation which can ensure the
filling strength and prevent particles creeping. For open hole
gravel packing operation, gravel size should be one magnitude
smaller.
5. Conclusions
A new gravel size design method named “HC&DF” for hy-
drate exploitation wells has been proposed in this paper.
Minimum gravel size to ensure discharge of fine component
in the formation was calculated first, and appropriate gravel
size for retaining coarse component can be calculated. Inter-
section of the above two datasets is the proper gravel size for
fluid extraction method in hydrate exploitation wells. Consid-
ering about layered sorting, shale content, etc., the design
result is matched with industrial gravel size to meet applica-
tion request.
Site X located in Shenhu area, Northern South China Sea
was taken as an example to illustrate detailed gravel sizing
procedure based on HC&DF method. The research result
suggests that gravel size for site X should be 215360 μm.
As described above, All fine components should be dis-
charged during HC&DF method, so proper fluid infusion into
the wellbore and sand carrying working system are very im-
portant for gravel packing wells. Fine control of gas, liquid
and solid phases lead to effective and long-term hydrate ex-
ploitation operation.
Nomenclature
C—uniformity coefficient of formation sand, dimensionless;
d—particle size of formation sand, m;
d5, d10, d16, d40, d84, d90, d95—particle diameter when cumulative
 1020 
mass/volume is 5%, 10%, 16%, 40%, 84%, 90% and 95% on PSD
curve, μm;
d50—grain median size, μm;
dfmax—the maximum particle size of fine component, μm;
Dcmax, Dcmin—the maximum gravel size and minimum gravel size
for retaining coarse component, respectively, μm;
Dfmin—minimum gravel size for discharging fine component with
shale content considered, μm;
'
Dfmin —minimum gravel size for discharging fine component
without considering shale content, μm;
F—sorting coefficient, dimensionless;
M—total sand mass used for particle size analysis, kg;
N—total particle number used for particle size analysis;
Rm—loss ratio during sand retention caused by shale content,
160%;
W—mass fraction of formation sand at a given size, %;
ρ—density of formation sand, kg/m3.
References
[1] WU Nengyou, ZHANG Haiqi, YANG Shengxiong, et al. Gas
hydrate system of Shenhu area, Northern South China Sea:
Geochemical results. Journal of Geological Research, 2011,
2011: 110.
[2] ZHANG G, LIANG J, LU J, et al. Geological features, con-
trolling factors and potential prospects of the gas hydrate oc-
currence in the east part of the Pearl River Mouth Basin, South
China Sea. Marine & Petroleum Geology, 2015, 67: 356367.
[3] GUO Xiangyu, LIU Fei. China announced combustible ice
test success in the Shenhu area, South China Sea. (2017-
05-18)[2017-06-01]. http://china.cnr.cn/xwwgf/20170518/
t20170518_523762235.shtml.
[4] CHEN Huiling, ZHU Xia. 20 years to catch up and 60 days to
break: Catching up recording of China’s coastal combustible
ice exploration test. (2017-07-17)[2017-07-26]. http://www.
cgs.gov.cn/ddztt/jqthd/trqshw/zxbdshw/201707/t20170717_43
5792. html.
[5] LI Yanlong, LIU Changling, LIU Lele. Damage statistical
constitutive model of hydrate-bearing sediments and the de-
termination method of parameters. Acta Petrolei Sinica, 2016,
37(10): 12731279.
[6] JUNG J W, JANG J, SANTAMARINA J C, et al. Gas produc-
tion from hydrate-bearing sediments: The role of fine parti-
cles. Energy & Fuels, 2012, 26(1): 480487.
[7] YOSHIHIRO T, DUNCAN M W, HAY W J, et al. Deepwater
methane hydrate gravel packing completion results and chal-
lenges. OTC 25330-MS, 2014.
[8] LI Yanlong, LIU Lele, LIU Changling, et al. Sanding predic-
tion and sand-control technology in hydrate exploitation: A
review and discussion. Marine Geology Frontiers, 2016,
32(7): 3643.
[9] LIU Changling, MENG Qingguo, HU Gaowei, et al. Charac-
terization of hydrate-bearing sediments recovered from the
Shenhu area of the South China Sea. Interpretation, 2017,
 LI Yanlong et al. / Petroleum Exploration and Development, 2017, 44(6): 1016–1021
5(3): 1323.
[10] DENG Jingen, LI Ping, ZHOU Jianliang, et al. Sand control
optimization applied to moderately sanding wells in offshore
loose sandstone reservoirs. Acta Petrolei Sinica, 2012, 33(4):
676680.
[11] DENG Jingen, LI Ping, WANG Lihua, et al. The optimization
of sand control method for moderate sand control technique
application in Bohai Bay. Oil Drilling & Production Technol-
ogy, 2011, 33(1): 98101.
[12] WANG Lihua, DENG Jingen, ZHOU Jianliang, et al. Experi-
mental study on premium screen mesh opening design for
reasonable sand control. China Offshore Oil and Gas, 2011,
23(2): 107110.
[13] HU Caizhi, PEI Bolin, LI Xiangfang, et al. Laboratory re-
search on plugging mechanism in gravel packed well. Journal
of the University of Petroleum,China (Edition of Natural Sci-
ence), 2004, 28(3): 4042.
[14] VALDES J, SANTAMARINA J. Particle clogging in radial
flow: Microscale mechanisms. SPE Journal, 2006, 11(11):
193198.
[15] ZHANG Wei, LIANG Jinqiang, LU Jing’an, et al. Accumu-
lation features and mechanisms of high saturation natural gas
hydrate in Shenhu Area, northern South China Sea. Petroleum
 1021 
Exploration and Development, 2017, 44(5): 670680.
[16] DONG Changyin, ZHANG Qinghua, GAO Kaige, et al.
Screen sand retaining precision optimization experiment and a
new empirical design model. Petroleum Exploration and De-
velopment, 2016, 43(6): 991996.
[17] OYENEYIN B. Developments in petroleum science. Amster-
dam: Elsevier, 2015: 191223.
[18] DONG Changyin. Sand control theories and technologies in
oil and gas wells. Dongying: Press of China University of Pe-
troleum, 2012: 9197.
[19] MARKESTAD P, CHRISTIE O, ESPEDAL A, et al. Selection
of screen slot width to prevent plugging and sand production.
SPE 31087, 1995.
[20] SAUCIER R J. Consideration in gravel pack design. Journal
of Petroleum Technology, 1974, 26(2): 205212.
[21] DONG Changyin, JIA Bixia, LIU Chunmiao, et al. Blocking
mechanism and blocking laws experiments of sand retention
media in mechanical screens. Journal of China University of
Petroleum (Edition of Natural Science), 2011, 35(5): 8288.
[22] MA Shuai, XIONG Youming, YUDong, et al. Research on
precision design of sand control on high yield offshore gas
field. Oil Drilling & Production Technology, 2013, 35(6):
4851.