You are on page 1of 10

Powder Technology 312 (2017) 184193

Contents lists available at ScienceDirect

Powder Technology

journal homepage: www.elsevier.com/locate/powtec

On uniaxial compression and Jenike direct shear testings of cohesive iron


ore materials
Wei Chen , Alan Roberts, Kenneth Williams, Jeremy Miller, Jens Plinke
Centre for Bulk Solids and Particulate Technologies, The University of Newcastle, Callaghan 2308, Australia

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

Article history: The ow function of an iron ore material governs its owability characteristics in material handling chains of the
Received 9 November 2016 resource industry. A uniaxial compression test is able to obtain a ow function more rapidly compared to the
Accepted 15 February 2017 Jenike direct shear test, nevertheless, results often exhibit lower rankings using the former method. This study
Available online 20 February 2017
aims to investigate the fundamental stress states within the test specimen that led to this phenomenon, and to
introduce a new uniform density specimen preparation method for a uniaxial compression test in order to
Keywords:
Flow function
achieve comparative ow functions as per the Jenike direct shear test. The minimisation of the wall friction effect
Uniaxial compression test and the achievement of the critical state when preparing a uniaxial specimen were explicitly discussed. Experi-
Jenike direct shear test mental investigations on ow functions of a suite of Australian iron ore samples were conducted using both the
Iron ore uniform density uniaxial compression test and the Jenike direct shear test. Results from both methods were in-
Compressibility index dicated to be comparable providing the specimen exhibited cohesive ow behaviours. Additionally, a simple
compressibility index, based on the bulk density test of iron ore samples, was derived as a threshold to indicate
if a uniform density uniaxial compression test can produce ow functions matching the Jenike direct shear test.
The outcome of this research enabled a rapid and reliable ow function testing method for cohesive iron ore
materials.
2017 Elsevier B.V. All rights reserved.

1. Introduction Nevertheless, the Jenike direct shear test is, of necessity, rather time
consuming where the aim for reliable and reproducible results, requires
The rapid consumption of near surface iron ore deposits in Australia an experienced operator to perform the pre-consolidation, pre-shear
has led to mining deeper deposits which are located close or even be- and shear procedures, from which a ow function is obtained (as
neath the water table [1]. The resulting increase of the inherent mois- demonstrated in Fig. 1). For more efcient owability monitoring of co-
ture for the run-of-mine material leads to more cohesive and adhesive hesive iron ore materials, the requirement for a simpler, more rapid
behaviours, which cause poor owability in material handling chains testing method has a high priority.
[2]. Therefore, it is an industry standard to monitor the owability of Based on the foregoing objective, the uniaxial compression test rep-
bulk materials to minimise potential blockages. The owability of bulk resents a potential method to obtain ow functions [5,6]. Its simplicity
materials is governed by the ow function, which is a correlation be- and shorter testing time are often preferred in industrial practice [7].
tween the unconned yield strength with respect to the major principle As shown in Fig. 2(a), in a conventional uniaxial compression test, the
stress [3]. Among various testing methods devised to measure the ow sample is poured into a cylindrical mould and consolidated under a
function, the Jenike direct shear test (JDST) is widely accepted; and this pre-determined normal stress 1. The applied load corresponding to
test is particularly relevant to the design and efcient operation of bulk the consolidating stress is then removed followed by the careful
solids storage and handling systems for an extensive range of industries, retraction of the cylindrical mould to leave a free standing, consolidated
such as those involved in the mining and processing of iron ore [4]. cylindrical test sample without lateral constraint. The sample is then
subjected to an increasing normal compressive stress until failure oc-
curs. The normal stress at failure is deemed to be the unconned yield
strength c. The stress c corresponding to the consolidation stress 1
Corresponding author at: Centre for Bulk Solids and Particulate Technologies,
Newcastle Institute for Energy and Resources, The University of Newcastle, Callaghan
denes one point on the ow function graph. The test is repeated for
2308, Australia. at least two other consolidation stresses to obtain a ow function.
E-mail addresses: W.Chen@newcastle.edu.au, w.chentbs@gmail.com (W. Chen). (See Fig. 3.)

http://dx.doi.org/10.1016/j.powtec.2017.02.037
0032-5910/ 2017 Elsevier B.V. All rights reserved.
W. Chen et al. / Powder Technology 312 (2017) 184193 185

direct shear test, the specimen was largely not affected by the wall fric-
tion effect due to the material being compacted into relatively thin
layers, which resulted the specimen in the stress state of HS(0), with
S(0) on the critical state line on the Roscoe surface and normal stress
of (0) and voids ratio of e(0). Whereas, in a conventional uniaxial com-
pression test, the effective normal stress applied to a test specimen was
reduced to (1) due to the wall friction effect, which resulted the speci-
men shifting from HS(0) to HS(1) stress state. This effectively leads to
higher voids ratio - e(0) and lower yield locus comparing to the stress
state of a Jenike direct shear test specimen.
Previous studies have attempted to overcome the wall friction effect
in the uniaxial compression test. Some successes were achieved
employing mathematical procedures to correct the uniaxial ow func-
tions [12,13]. Nevertheless, the corrected factor often varied according
to the material type, thus no unied theory was developed. Maltby
and Enstad [14] adopted the triaxial test specimen preparation method
Fig. 1. Flow function derivation from Jenike direct shear tests.
by wrapping a membrane around the sample and adding lubrication be-
tween the membrane and mould wall to minimise the wall friction ef-
fect. Alternatively, an innite layer specimen preparation method was
Since 1 and c are determined directly, the test is simpler to per- utilised attempting to eliminate the wall friction effect [15]. Both
form and less time consuming than the Jenike test which requires the methods resulted ow functions approaching the Jenike direct shear
values of minor principle stress 2 and unconned yield strength c to testing result. However, the testing procedure was rather complex
be derived indirectly from the yield loci as illustrated in Fig. 1. In the and, from a practical point of view, rather inefcient.
case of bulk materials of quite low cohesive strength, the uniaxial test While the wall friction effect may be minimised through the exper-
requires minimum consolidation level to be applied to the specimen imental strategies discussed above, the specimen may still fail to reach
to ensure the sample remains intact under the inuence of the gravita- the critical state if the critical voids ratio is not achieved. For the Jenike
tional forces after removing the mould. Most critically, the ow func- direct shear test, the limited travel of the top shear ring relative to the
tions obtained through a uniaxial test often exhibited lower rankings xed base of the shear cell usually requires a series of applied twists of
when comparing to the Jenike direct shear test [8]. This is illustrated the cell lid carrying the applied normal load as an initial phase of the
in Fig. 2(b) which compares the ow functions of an iron ore material shear consolidation of the contained sample. The aim is to ensure the
obtained through both methods. For the uniaxial tester, the problem sample reaches the critical voids ratio (e0) at the critical state HS(0).
centres around the wall friction in the mould, often referred to as the This is illustrated in Fig. 4 for the Hvorslev-Roscoe surfaces, without
Janssen effect [9] causing non uniformity of the major consolidation such twisting action to induce particle reassembly, a higher voids
stress 1 which reduces exponentially with respect to the specimen ratio (e1) was often obtained leading to lower shear strength on HS(1),
depth. This has been attributed to the underestimation of the ow func- which occurred in a conventional uniaxial compression test. This phe-
tion using the uniaxial compression test. nomenon was also often observed in a bulk density compressibility
Based on the foregoing comments, the purpose of this paper is to test. When a sample was compacted with twisting axial load compar-
critically evaluate the uniaxial test in relation to the Jenike direct shear ing to non-twisting axial load, higher bulk density was obtained using
test with aim of establishing a sample preparation procedure to achieve the former method. Therefore, apart from minimisation of the wall fric-
the necessary critical consolidation state for the uniaxial compression tion effect, it was also important to ensure the critical voids ratio within
test to ensure the validity of the ow function determination. a test specimen to be achieved in a uniaxial compression test, without
which the shear strength of a uniaxial specimen remained comparative-
2. Stress states in test specimens ly lower.
Based on the above discussion, the critical voids ratio of the test spec-
The inuence of the wall friction effect on the uniaxial compression imen was suggested to be more crucial in determining the shear strength
test outcomes can be investigated through Hvorslev and Roscoe sur- of the sample, which was not addressed in the conventional uniaxial
faces [10,11]. As demonstrated in Fig. 3, the Hvorslev surface dened compression test method. This research aims to adopt a uniform density
the shear strength of a specimen in the three dimensional space specimen preparation method to achieve the critical state of a sample. A
consisting of the voids ratio, normal stress and shear stress. In a Jenike suite of experimental investigations will then be conducted to examine

Fig. 2. Uniaxial compression test. (a) conventional uniaxial compression testing process; (b) typical ow function comparison between two testing methods.
186 W. Chen et al. / Powder Technology 312 (2017) 184193

Fig. 3. Analysis of the wall friction effect on the stress state of a uniaxial specimen comparing to the Jenike direct shear test specimen on the Hvorslev and Roscoe surfaces.

the performance of the proposed specimen preparation method on com- preparing the uniaxial compression test specimens. For equal height
paring ow functions to the Jenike direct shear method. for each layer, based on a predetermined specimen bulk density and
particle properties, the accumulative mass of material for layer - i was
determined using the following equation.
3. Uniform density specimen preparation method

   
In order to achieve the required uniform density in the uniaxial test m0  D2 H Un0 nall ni
Accumulative mass d 1 ni 1
sample, a dened pre-consolidation procedure needs to be established 100 4nall 100 nall 1
and included as an integral part of the sample preparation. The
undercompaction specimen preparation method [16] is the basis for where Un0 was the percent undercompaction for the rst layer (5%); ni
the procedure adopted in this study in order that a uniaxial specimen was the number of layer - i; nall was the total number of layers; d was
to reach the comparable compaction state to that of a Jenike direct the dry bulk density; m0 was the initial moisture content; D and H are,
shear test sample. Rather than targeting a normal stress (1) during respectively, the diameter and height of the test specimen. The required
the sample preparation, in this case the method targets a nal bulk mass of the material for each layer was placed into a split mould and
density for the test specimen. Therefore, it is rstly necessary for a compacted to the required height with a tamping rod. The sample sur-
bulk density test utilising the Jenike compressibility tester [17]. face between layers was scratched prior to preparing the next layer.
Once a targeted specimen bulk density is selected, the sample is The specimen was fully formed after 10 sub lots of material were
compacted with a pre-dened number of layers. An optimal percentage compacted into a sample mould.
of pre-consolidation is applied to the rst (bottom) layer. This arrange-
ment ensures the achievement of a uniform density along the sample 4. Experimental scheme
height. When a test specimen is compacted in layers, the compaction
of each succeeding layer can further densify the sample below it. This Following the above procedure, the resulting bulk density of the ob-
was overcome using the undercompaction principle. As illustrated in tained specimen corresponded to a sample prepared using the similar
Fig. 5, each layer was typically compacted into a lower density than twisting action in a Jenike direct shear test. To examine the inuence
the nal desired value by a predetermined percent of undercompaction of the uniform density sample preparation method on ow functions
Un. The Un value in each layer was linearly varied from the bottom to the obtained from the uniaxial compression test, a comparative experimen-
top layer, with the bottom layer having the maximum Un value. tal program with the Jenike direct shear test was performed on a suite of
By way of background, research involving many cyclic triaxial tests iron ore samples.
performed on moist coarse grained soil samples, [16,18], showed A total of four iron ore samples were selected for the experimental
that maximum shear strength was achieved when utilising a 5% investigation. All samples were obtained through a series of crushing
undercompaction for the rst (bottom) layer and a total number of and screening from the onsite operation. Materials were subsequently
ten layers. These settings were adopted in this iron ore study for homogenised and sieved down to 4 mm size fraction. Where bulk
W. Chen et al. / Powder Technology 312 (2017) 184193 187

Fig. 4. Analysis of the critical voids ratio on the stress state of a uniaxial specimen comparing to the Jenike direct shear test specimen on the Hvorslev and Roscoe surfaces.

solids composed of particles of a large size range from coarse to nes, it were tabulated in Table 1. The particle size distribution of the samples
is the ne particles that contribute to the material's cohesive strength was also shown in Fig. 6.
[19]. Each sample was then prepared at moisture contents covering For each sample, a bulk density test was initially conducted follow-
the nominal operational range. The material properties for all materials ing the ASTM standard [17]. Results for all four samples at different

Fig. 5. Principle of the uniform density specimen preparation method.


188 W. Chen et al. / Powder Technology 312 (2017) 184193

Table 1 moisture contents were shown in Fig. 7. A suite of Jenike direct shear
Material properties for selected iron ore samples. tests was then conducted on each sample with a 101.6 mm outer
Sample label Particle density kg/m3 Moisture content diameter (4) shear cell following the ASTM standard [20]. In terms of
IO-A 4600 4.9%
the uniaxial compression test, both the conventional and the uniform
6.5% density specimen preparation methods were used on each sample for
7.9% comparative analysis. On the uniform density uniaxial compression
9.9% test (UDUCT), ve different bulk density values corresponding to a
IO-B 4400 9.2%
wide range of normal stresses were selected to undertake the test and
10.7%
13.7% to derive the ow function.
16.8% The uniaxial shear tester utilised in this research was shown sche-
IO-C 4200 12.1% matically in Fig. 8. The tester was designed to accommodate an
14.9% 80 mm in diameter and 160 mm in height specimen. Once the sample
18.5%
20.6%
was prepared, the conning walls were then retracted, after which the
IO-D 4300 9.5% top loading disc was driven by the pneumatic actuator to compress
11.6% the specimen. The resulting unconned yield stress was measured.
13.6% Three tests were performed on each sample.

5. Results and discussion

5.1. Flow function comparisons

Comparative ow functions of the uniaxial compression tests using


the conventional and uniform density sample preparation methods
were initially obtained and compared in Fig. 9. Using the conventional
sample preparation method, sample IO-B at 9.2% and IO-C at 12.1% col-
lapsed under the gravity force; therefore, no ow functions were ob-
tained. Whereas, all samples prepared using the uniform density
method were able to stand the gravity force. Comparing ow functions
obtained in both tests, it was observed that higher ranking was achieved
using the uniform density sample preparation method across all sam-
ples. Therefore, it was suggested that shear strength was enhanced
when samples were prepared using the uniform density method.
Jenike direct shear test results were also obtained and subsequently
compared to ow functions from the uniform density uniaxial compres-
Fig. 6. Particle size distribution of all samples. sion tests. Fig. 10 demonstrated the comparative results between the

Fig. 7. Bulk density test results for all samples.


W. Chen et al. / Powder Technology 312 (2017) 184193 189

matched tests, the sample exhibited increasing shear strength when


the compression load was initially applied. After a peak stress was
achieved, the sample demonstrated some level of residual stress when
continued the compression loading. Shear planes were clearly observed
after the failure of the sample. In comparison, for unmatched tests, the
sample quickly collapsed after showing some levels of shear strength.
No clear shear planes were observed and free owing behaviours
were indicated. Such phenomenon was observed throughout all iron
ore samples tested.
The ow function of a bulk material can also be categorised based on
the Jenike owability classication [5], which was dened as

1
ff c 2
c

where 1 was the major consolidation stress and c was the unconned
yield strength. ffc characterised the following ow behaviours,
ffc N 10: Free owing, very low cohesion
4b ffc b 10: Easy owing, low cohesion
2b ffc b 4: Cohesive
1b ffc b 2: Very cohesive
ffc b 1: Not owing

All ow functions from the uniform density uniaxial compression


test were plotted against the Jenike owability classication in
Fig. 12. Matched and unmatched ow functions to the Jenike direct
shear test were differentiated. It was clearly observed that all
matched ow functions from the uniform density uniaxial compres-
sion test were scattered in the cohesive (2 b ffc 4) and very cohesive
(1 b ff c 2) regions. In comparison, the unmatched ow functions
were distributed in the free owing, very low cohesion (ff c N 10)
and the easy owing, low cohesion (4b ffc 10) regions. Consequently,
it was suggested that, for iron ore materials, there existed an appar-
ent cohesion threshold above which the uniform density uniaxial
compression tests were able to produce comparable ow functions
to the Jenike direct shear test. Under such cohesion level, the speci-
men exhibited low shear strength, therefore, free owing behaviours
were induced.
Additionally, it was indicated that a single Jenike owability
index ffc was not suitable to dene the overall owability of an iron
ore material, since a ow function often spanned over two classica-
tions depending on the major consolidation stress (1) level. From a
practical perspective, the Jenike owability classication was
Fig. 8. Image and schematic of the uniaxial compression tester used in this study. sensible in the low consolidation stress region where the mass
ow regime dominates. However, at high consolidation stress,
the linear extrapolation of the Jenike owability classication
appeared to under-estimate the material handling difculties since
the ow function tend to plateau and funnel ow is commonly
two tests. It was indicated that the ranking of the ow functions be-
observed for cohesive iron ore material under high consolidation
tween the two tests matched well for samples with relatively higher
stress [21].
moisture contents, including

IO-A at 6.5%, 7.9% and 9.9%; 5.2. Indication of cohesive thresholds based on conned compressibility
IO-B at 13.7% and 16.8%;
IO-C at 18.5% and 20.6%; The bulk density curves for all iron ore samples also showed correla-
IO-D at 11.6% and 13.6%. tions with whether a uniform density uniaxial ow function was able to
match the corresponding Jenike test ow function. As shown in Fig. 7,
the bulk density curves for all ow function matched test cases
However, at lower moistures, the uniform density uniaxial compres- appeared to span in larger bulk density ranges comparing to the un-
sion test still led to under-estimation of the ow functions comparing to matched test cases. This was particularly evident for sample IO-B and
the Jenike direct shear test. Further investigation on the specimen IO-C. The bulk density curve essentially reected the compressibility
failure behaviours indicated that these unmatched tests failed to of an iron ore sample, and can be dened as
achieve a peak unconned yield stress during the compression loading.
Fig. 11 demonstrated the distinct axial stressaxial strain relationship
during compression loading for both a ow function matched test case B 1 B 0
3
(Fig. 11(a)) and a ow function unmatched test case (Fig. 11(b)). In B 0
190 W. Chen et al. / Powder Technology 312 (2017) 184193

Fig. 9. Comparative ow function results obtained through conventional uniaxial compression test (UCT) and the uniform density uniaxial compression test (UDUCT) for all samples.

where B(1) and B(0) were the bulk density under 1 and zero normal where V(1) and V(cell) were the sample volume under a normal stress
stresses. Since the material mass input into a bulk density test cell was of 1 and the bulk density test cell volume. Therefore, Eq. (3) can be
xed, therefore, transformed to
V 1
1
B 1 B 0 Vcell Vcell
B 1 1: 5
B 1 V 1 B 0Vcell 4 B 0 V 1 V 1
Vcell

Fig. 10. Comparative ow function results obtained through the Jenike direct shear test (JDST) and uniform density uniaxial compression test (UDUCT) for all samples.
W. Chen et al. / Powder Technology 312 (2017) 184193 191

Fig. 11. Distinct axial stress-axial strain behaviours in the uniform density uniaxial compression tests. (a) ow functions matched test cases; (b) ow functions unmatched test cases.

threshold exceeding which the uniform density uniaxial compression


test was able to produce comparable ow functions to Jenike direct
shear test.
The variation of the compressibility of an iron ore sample at different
moistures was predominantly due to the clay minerals, such as Kaolin,
contained in the material [22]. An important characteristic of the clay
content in the iron ore material was inducing the macro shrink-swell
behaviours. At moist state, the sample tended to swell, thus producing
higher volume change under a normal load. Conversely, at dry state,
shrink behaviour was exhibited, resulting smaller volume change
under a normal consolidation stress.
In the meantime, the cohesive behaviour of the iron ore sample was
also due to the combined clay-moisture effect [23]. For an iron ore sam-
ple, the apparent cohesion emerged as soon as the material was moist-
ened to a certain degree. This often led to the overall enhancement of
Fig. 12. Distribution of the ow functions obtained through uniform density uniaxial the shear strength when adding more water to the sample. As previous-
compression tests (UDUCT) in the Jenike owability classications.
ly discussed, the uniform density uniaxial compression test was only
able to produce comparable ow functions to the Jenike direct shear
Since an identical cross-sectional area applied to V(1) and V(cell), test when cohesive behaviours were exhibited by the iron ore sample.
Eq. (5) can be expressed as, Therefore, based on the above analysis, it was suggested that the
compressibility can be utilised as a threshold to indicate the apparent
cohesive ow behaviour of an iron ore sample, since the two macro
Hcell
1 6 properties of an iron sample due to the combined effect of the clay con-
H 1
tent and moisture were cohesion and compressibility.
where H(cell) and H(1) were cell height and the sample height under The logarithmic like axial strain-axial stress correlations shown in
normal stress of 1. was also referred as the axial strain in the conned Fig. 13 can be converted to quasi-linear relationships as shown in
compression test. Therefore, it was viable to transform all bulk density Fig. 14. By this means, the compressibility of a sample can be simply
curves into an axial strain-axial stress chart as shown in Fig. 13. From
the gure, it was clearly indicated that there existed a compressibility

Fig. 13. Correlations between normalised axial strain and axial stress for all samples. Fig. 14. Quasi-linear correlation between the axial strain to axial stress.
192 W. Chen et al. / Powder Technology 312 (2017) 184193

Table 2
Quasi-linear parameters for the twisted axial strainaxial stress correlations.

R2 Sample Matched between


log10 1 UDUCT and JDST

0.058 0.99 IO-C-12.1% N


0.065 0.99 IO-B-9.2%
0.094 0.99 IO-B-10.7%
0.099 0.99 IO-A-4.9%
0.101 0.98 IO-C-14.9%
0.104 0.99 IO-D-9.5%
0.140 0.98 IO-D-11.6%
0.152 0.99 IO-A-6.5% Y
0.195 0.99 IO-A-7.9%
0.201 0.99 IO-D-13.6%
0.212 0.95 IO-B-13.7%
0.239 0.98 IO-C-18.5%
0.293 0.94 IO-A-9.9%
0.307 0.99 IO-C-20.6%
0.342 0.99 IO-B-16.8%

represented by the slope of the linear tting for axial strain-logarithmic compressibility and cohesion properties induced by the combined ef-
axial stress correlation, which was dened as, fect clay and moisture content within the sample.

Consequently, a uniform density uniaxial compression test can be



: 7 adopted during iron ore mining operation for rapid and reliable
log10 1
owability indications. Once implemented, this method is able to
increase the efciency and reduce potential blockages of the material
All tting function parameters were tabulated in Table 2. Results handling plants.
indicated that the uniform density uniaxial compression test on an
iron ore sample was able to produce comparable ow functions to the
Jenike direct shear test, providing the compressibility log 1 of the sam- Acknowledgements
10

ple obtained in the bulk density test exceeds 0.152. Practically, such em-
pirical compressibility index can be utilised to indicate the cohesive The authors are grateful for the nancial assistance of the ARC
ow behaviours of the iron ore material by simply conducting a bulk Research Hub for Advanced Technologies for Australian Iron Ore
density test. A simple and fast uniform density uniaxial compression (IH130200031).
test can be subsequently performed if quantitative assessment on the
owability of the iron ore material is required. The shorter turnaround References
time based on this arrangement for owability monitoring is able to re- [1] Geoscience Australia, Sustaining the Mineral Resources Industry Overcoming the
duce the material handling blockages during the iron ore mining Tyranny of Depth, 2008.
operation. [2] J. Plinke, J.-D. Prigge, K.C. Williams, Development of new analysis methods for the
characterization and classication of wet sticky ores, Powder Technol. 294 (2016)
252258.
6. Conclusion [3] A.W. Jenike, Storage and ow of solids, bulletin no. 123, Bull. Univ. Utah 53 (26)
(1964).
[4] A.W. Roberts, O.J. Scott, An investigation into the effects of sinusoidal and random
A comprehensive investigation was carried out on the comparative
vibrations on the strength and ow properties of bulk solids, Powder Technol. 21
ow functions of cohesive iron ore samples obtained using the uniform (1) (1978) 4553.
density uniaxial compression test and the Jenike direct shear test. A [5] D. Schulze, Flow Properties of Powders and Bulk Solids, Braunschweig/Wolfenbu
ttel, Ger. Univ. Appl. Sci., 2006
suite of experimental tests on various iron ore samples were performed,
[6] A. Kwade, D. Schulze, J. Schwedes, Determination of the stress ratio in uniaxial
and analysis on these results were presented. The study yielded the compression tests, Powder Handl. Process. 6 (1) (2013) 61.
following major ndings: [7] D. Schulze, Powders and Bulk Solids, Behav. Charact. Storage Flow, Springer, 2008.
[8] J. Schwedes, Review on testers for measuring ow properties of bulk solids, Granul.
A uniform density sample preparation method ensured the uniform Matter 5 (1) (2003) 143.
and critical compaction in uniaxial compression test specimens were [9] R.M. Nedderman, Statics and Kinematics of Granular Materials, Cambridge University
Press, 2005.
comparable to Jenike direct shear test specimens. [10] K.H. Roscoe, A. Schoeld, C.P. Wroth, On the yielding of soils, Geotechnique 8 (1)
Uniform density sample in uniaxial compression tests signicantly (1958) 2253.
enhanced the shear strength of specimens, and thus, producing [11] M.J. Hvorslev, ber die Festigkeitseigenschaften gestrter bindiger Bden: With an
abstract in English, no. 45, Danmarks naturvidenskabelige samfund, i kommission
improved ow functions comparing to the conventional uniaxial has GEC Gad, 1937.
compression test. [12] S. Wiche, A. Roberts, W. McBride, Development and Application of a Uniaxial
Flow functions of iron ore samples obtained using the uniform density Type Flowability Tester for Characterising Bulk Solids, Particulate Systems Anal-
ysis, 2005.
uniaxial compression test were comparable to the Jenike direct shear
[13] L. Parrella, D. Barletta, R. Boerejn, M. Poletto, Comparison between a uniaxial
test ow function when the material exhibited apparent cohesive compaction tester and a shear tester for the characterization of powder owability,
ow behaviours. Without sufcient cohesion stress, specimen exhib- KONA Powder Part. J. 26 (0) (2008) 178189.
[14] L.P. Maltby, G.G. Enstad, Uniaxial tester for quality control and ow property
ited free owing behaviours, which resulted to lower ow function
characterization of powders, Bulk Solids Handl. 13 (1993) 135.
rankings to Jenike direct shear test ow functions. [15] J.C. Williams, A.H. Birks, D. Bhattacharya, The direct measurement of the failure
The compressibility of an iron ore sample obtained through a bulk function of a cohesive powder, Powder Technol. 4 (6) (1971) 328337.
density test can be utilised to indicate if a uniform density uniaxial [16] R.S. Ladd, Preparing Test Specimens Using Undercompaction, 1978.
[17] ASTM International, ASTM D6683-14, Standard Test Method for Measuring Bulk
compression test can produce ow functions matching the Jenike Density Values of Powders and Other Bulk Solids as Function of Compressive
direct shear test. This was due to the correlation between the Stress(West Conshohocken, PA) 2014.
W. Chen et al. / Powder Technology 312 (2017) 184193 193

[18] M. Vucetic, R. Dobry, Cyclic triaxial strain-controlled testing of liqueable sands, [21] A. Roberts, Handleability or Flowability, Centre for Bulk Solids and Particulate
Advanced Triaxial Testing of Soil and Rock, ASTM International, 1988. Technologies, The University of Newcastle, Australia, 1999.
[19] A. Roberts, Basic Principles of Bulk Solids Storage, Flow and Handling, Institute for [22] H.H. Murray, Traditional and new applications for kaolin, smectite, and palygorskite:
Bulk Materials Handling Research, 1993. a general overview, Appl. Clay Sci. 17 (5) (2000) 207221.
[20] ASTM International, ASTM D6128, Standard Test Method for Shear Testing of Bulk [23] E. oka, O. Erol, F. Armangil, Effects of compaction moisture content on the shear
Solids Using the Jenike Shear Cell, ASTM International, West Conshohocken, PA, strength of an unsaturated clay, Geotech. Geol. Eng. 22 (2) (2004) 285297.
2015.

You might also like