IBP1496_19

DESIGN AND COMMISSIONING OF A LONG-

DISTANCE PIPELINE TRANSPORTING THICKENED
REJECTS IN LAMINAR FLOW
Alisson M. Vidal1, Rafael C. Lima2,
Jay P. Chapman , Charleston Dias4, Daniel R. de Sá5.
3

Lima5,
Copyright 2019, Brazilian Petroleum, Gas and Biofuels Institute - IBP
This Technical Paper was prepared for presentation at the Rio Pipeline Conference and Exhibition 2019, held
between 03 and 05 of September, in Rio de Janeiro. This Technical Paper was selected for presentation by the
Technical Committee of the event according to the information contained in the final paper submitted by the
author(s). The organizers are not supposed to translate or correct the submitted papers. The material as it is
presented, does not necessarily represent Brazilian Petroleum, Gas and Biofuels Institute’ opinion, or that of its
Members or Representatives. Authors consent to the publication of this Technical Paper in the Rio Pipeline
Conference and Exhibition 2019.

Abstract

Transportation of mineral slurry through pipeline is widely used in projects around the
world. However, for most operations the transport flow is turbulent, where a relatively high
transport velocity is required to hold all particles suspended, with size characteristics of 100
micron and 45 to 85% <45 microns.
In this work, the design and commissioning results of a thickened bauxite tailings
pipeline using laminar transport flow will be discussed. This pipeline is a 24-inch outside
diameter and 8-kilometer length located in the Northern Region of Brazil. During the project
development, laboratory tests were carried out with the tailings and the main characteristics
were: size of 100% <300 microns and 99% <45microns, at a concentration of 35% solids by
weight, specific density of 2.06 t / m³, viscosity and yield stress of approximately 50 cP and
100 Pa, respectively.

Keywords: Bauxite Tailings, Slurry Pipeline, Laminar Flow

1. Introduction

To produce 9.9 million tons per year of bauxite, a mining company located in northern
region of Brazil had two independent Beneficiation Plants. The tailings from the cycloning
overflow were thickened and pumped to the tailings DAMs.
The previous pumping consisted of three (3) centrifugal pumps, installed in two (2)
thickeners underflows, and then pumped to two pump trains, consisting of 3 pumps each.
Due to the forecast of production increase from 9.9 to 14.9 million tons per year, it was
necessary to study and implement new DAMs, which required a new pumping system.
The new pumping route had two deployment phases:

1. 5.9 km main pipeline and 8.7 km of spillage in 2017;

______________________________
1
Lead Pipeline Engineer - Ausenco
2
Pipeline Director – Brazil - Ausenco
3
International Consultant - Global Practice Lead – Pipelines – Ausenco
4
Senior Engineer – Ausenco
5
Engineering Technical Auxiliar – Ausenco
 Rio Pipeline Conference and Exhibition 2019

2. 6.5 km main pipeline and 13.2 km of spillage in 2021.

The conceptual study initiated in laboratory tests, in order to define the material
physicochemical characteristics.

2. Laboratory Tests

The first step of the conceptual study was the bauxite tailings rheological tests from
each processing plant, performed by Ausenco do Brasil Engenharia Ltda, with the purpose of
providing the design data for the of hydraulic calculations development.
Core parameters were determined to give support to the project, such as:
a. Solids percentage and pH value by drying aliquots of the homogenized sample;
b. Particle size distribution using the Malvern Instruments Mastersizer 2000 Laser
Particle Size Analyzer;
c. Specific mass of the solid using Le Chatelier Balloon.

Three different methods for rheology analysis were performed: rotational rheometry
(Bingham and Casson model), Vane test and Slump test. The rheological tests were performed
using Haake VT 550 viscosimeter.

2.1 Rheological Calculation Methods

The Bingham Model determines the shear stress as a function of the yield stress, shear
rat and the slurry viscosity.
𝜏 = 𝜏𝑦 + 𝜂𝛾

Where  is the shear stress (Pa); 𝜏𝑦 is yield stress (Pa); 𝛾 is the shear rate (1/s) and 𝜂 the
slurry viscosity (Pa. s).

The Casson Model is similar to the Bingham Model.

√𝜏 = √𝜏𝑦 + √𝜂𝛾

For comparative purposes, it was also performed the Vane T, using a vane sensor
FL100 to determine the flow voltage.

The Slump test was performed using a PVC tube with 99 mm diameter and 98 mm
height, to determine deformity and the flow stress through the HALLBOM model presented
by the following equation:
(1 − 𝑠 ′ )
𝜏′ = ′
2𝑒 √3𝑠
Where:
𝜏𝑦
' is the dimensionless flow voltage given by: 𝜏 ′ = 𝜌𝑔𝐻;
𝑆
S ' is the dimensionless slump given by: 𝑠 ′ = 𝐻;
H is the height of the tube and S is the slump (abatement).

2
 Rio Pipeline Conference and Exhibition 2019

2.2 Results

Table 1 shows the solid concentration, pH value and the specific solids mass of the
tested sample.

Table 1 – Sample Properties

Concentration (Cw) Specific mass (Δs)
pH
(% solids, by weight) Dry solid (g/cm³)
32.96 5.4 2.80

Figure 1 presents the sample average particle size distribution, using the Mastersizer
2000 Laser Diffraction Analyzer.

Figure 1. Particle size distribution – Project sample

Figure 2 shows the yield stress values results, depending on the percentage of solids, for
all different models.

3
 Rio Pipeline Conference and Exhibition 2019

Figure 2. Flow rate in percentage function using rotational rheometer tests, Vane test and Slump test.

All rheology tests showed high yield stress values, influence of the low-grade particle
size, increasing the vehicle viscosity. It was observed that the Bingham model was more
conservative, showing the highest flow stress values, while the Vane test presented less
conservative with the lowest flow stress values Found.
For the hydraulic calculations it was used the Bingham model, because it presented the
most conservative scenario.

3. Hydraulic Calculation

The transition velocity calculations showed that the velocity needed to achieve the
turbulent regime was not feasible, making the implantation of the system impracticable. The
curve of the Reynolds number vs. system speed shown in Figure 3 shows the transition speed
is close to 25 m/s.

4
 Rio Pipeline Conference and Exhibition 2019

Figure 3. Reynolds Number vs. Transport Velocity

This project merged to supply the mining company need to pump tailings in a higher
concentration than it was currently being pumped. Due to the impossibility to reach a turbulent
flow without compromising the pipeline system, all hydraulic calculations were made
considering laminar flow.
The minimum transport velocity recommended in the conceptual study was 0.6 m/s, due
to the minimum flowrate required by the client.

3.1 Pump Selection

During the conceptual study, the pumping possibilities were analyzed using centrifugal
and positive displacement pumps.
The quantity of pumps for each layout studied presented in Table 4 presents a
comparison between the systems selected to perform in the operating conditions of the system.

Table 4. Pump quantities

Phase 2
Pump type Phase 1
Centrifugal 40 (26 operational + 14 reserve) 86 (63 operational + 23 reserve)
Positive Displacement 4 (3 operational + 1 reserve) 6 (5 operational + 1 reserve)

It was also analyzed the possibility of using a trunk line for the three plants. However,
this option has not become feasible for centrifugal pumps.
As showed in Figure 4, it is schematically that the expected performance of centrifugal
pumps operating with homogeneous slurry that present shear stress presents a change in shape.

5
 Rio Pipeline Conference and Exhibition 2019

Figure 4. Effect of high yield stress on pump performance (BOOTLE, 2002)

The bottleneck was the high variation of the flow stress for a small variation in the
percentage of solids.
In addition to the cost analysis for pumping using the two technologies, the operational
difficulties were also used, due to the laminar nature of the pumping.
The use of positive displacement pumps was defined for presenting better conditions of
system design, operation and mostly due to the slurry high rheology. Figure 5 shows the pump
station layout.

Figure 5. Pump station layout

The pumps selected for this system were six (6) Geho TZPM 2000, maximum flow of
636 m³/h@ 11,700 kPa, 3000 HP.
6
 Rio Pipeline Conference and Exhibition 2019

4. Commissioning

During the commissioning phase in 2017, new tests were carried out with the tailings,
to compare the conceptual study made 2015 with commissioning parameters.
Table 5 shows the solid concentration, pH value and solids density of the test samples
performing in the project and in the commissioning.

Table 5. Sample Properties Comparison

Sample Concentration (Cw) pH Specific mass (Δs)
(% solids, by weight) Dry solid (g/cm³)
Project 32.96 5.4 2.80
Commissioning 34.85 7.3 2.70

Figure 6 shows the particle size distribution for both samples, project and
commissioning.

Figure 6. Particle size distribution – Project and commissioning

Figure 7 shows the yield stress values according to the solids concentration of the
samples using the Bingham model. The used sample for the commissioning comparisons was
the AdoB547, due to be the sample that showed closest head loss to the real one observed.

7
 Rio Pipeline Conference and Exhibition 2019

Figure 7. Flow stress in function of the percentage of solids in the sample using the Bingham model.1

The slurry rheology variation found in laboratory tests with the first data of pumping
pressure showed that the Particle Size Variation in the passing percentage (%) in 10 µm
indicates different values for the flow stress, higher yield stress values were observed in the
samples with the highest accumulated percentage in the 10 µm mesh.
Figure 8 presents the Bingham yield stress graph in relation to percentage thru
accumulated in 10 µm.

Figure 8. Bingham flow stress in% passing ratio accumulated in 10 µm

8
 Rio Pipeline Conference and Exhibition 2019

The graphic showed in Figure 9 presents a discharge pressure comparison between

project, commissioning and actual pumping. The profile considered in this case is from the
pump station to one of the spikes installed in field.

Figure 9. Profile and Hydraulic Gradient

As showed in the graphic above, the pumping pressure were lower than the considered
in project phase. This difference can be explained by the differences between samples and its
high rheology variation in lower concentration changes, it is expected that distinctions may
occur.
During the commissioning phase, the pipeline was stopped with tailings inside for about
18 hours. The re-start did not show any abnormal conditions and the pump system was able to
achieve the required pressure to start pumping the tailings again.
After all tests were done, it was concluded that the minimum transport velocity could
be reduced from 0.6 to 0.4 m/s, without compromising the client expectations.

4. Conclusion

The commissioning phase was a success and confirmed what the project expectations.
The system is operations since 2017 without any pumping issues, showing itself to be an
economical and operational success for the client.
An attention point could be the chances of sedimentation inside the pipeline due to the
laminar flow, in some studies supported by AUDE, et al. 1996. Although the sedimentation
phenomena can not be avoided, the recommendation is to always monitor the measured
pressures and see if its increasing.
Beyond sedimentation issues, it is also part of a future study the pipe bottom wear due
to the possibility of coarse particles existence.

9
 Rio Pipeline Conference and Exhibition 2019

4. References

AUDE, T.C., DERAMMELAERE, R. H., WASP, E. J. Instability of Laminar Flow in Long

Distance Pipelines and Solutions. Florida, U.S.A. Coal Utilization & Fuel Systems 21st
International Technical Conference. mar., 1996.
AUSENCO. (2015). Laboratory Tests.
AUSENCO. (2015). Conceptual and Basic Project.
AUSENCO. (2017). Detailed Project / Commissioning.
BOOTLE, M. J. Selection and Sizing of Slurry Pumps. Wisconsin, U.S.A. Weir Slurry Group,
Inc. 2002.
HALLBOM, D.J., The “Lump” Test
VIII International Seminar on Paste and Thickened Tailings, Santiago, Chile, April 20-22, 2005
WASP, E.J.; KENNY, J.P.; GANDHI, R.L. Solid-liquid flow slurries pipeline transportation.
TransTech Publications, 1977.

10