Mining, Metallurgy & Exploration

https://doi.org/10.1007/s42461-021-00378-y

Tailings Filtration Using Viper Filtration Technology—a Case Study

Oliver Whatnall 1 & Kevin Barber 1 & Peter Robinson 2

Received: 28 September 2020 / Accepted: 5 January 2021

# The Author(s) 2021

Abstract
Investigation and uptake of filtered tailings continues to grow throughout the globe. This is driven by a wide range of site-specific
considerations, which include such factors as tailings characteristics (e.g., amenability to filtration), production rates, climate,
water availability, cost drivers, environmental requirements, and social factors. Despite the aforementioned technological growth,
the currently available filtration technology is not able to meet the needs of many operations and projects that would otherwise
adopt the technology. Experience with large-scale industrial filtration shows that vacuum belt filter systems meet the needs of
many modern users, exceptions being the inability to effectively dewater tailings at altitude and/or with a fine particle size
distribution: a potential fatal flaw. This paper presents a case study on the utilization of the patented Viper Filtration technology
on gold tailings to overcome this challenge and shares the resultant full-scale plant design, highlighting the features designed to
overcome cost and scalability deterrents. This technology is a novel mechanical process which complements the vacuum pressure
in dewatering the filter cake as it travels along the belt filter. This project commenced with a pilot testing program, which
successfully met the objective to rigorously test, measure and record any performance improvements achieved when engaging
the Viper technology. Of the two tailings products tested, gross improvements of 4.2%w/w and 5.7%w/w were achieved when
compared to the conventional vacuum belt filter operation. This pilot testing facilitated measurement of operating and design
data, which forms the basis of the full-scale system design and resultant equipment supply of three vibration roller assemblies for
retro-fitting on the existing vacuum belt filter.

Keywords Tailings . Dewatering . Filtration . Dry-stack . Gold . Viper

1 Introduction 1.1 Filtered Tailings

Experience with large-scale industrial filtration shows that Investigation and uptake of filtered tailings continues to grow
vacuum belt filter systems meet the needs of many modern throughout the globe. This is driven by a wide range of site-
users, exceptions being the inability to effectively dewater specific considerations, which include such factors as tailings
tailings at altitude and/or with a fine particle size distribution: characteristics (e.g., amenability to filtration), production
a potential fatal flaw. This paper presents a case study on the rates, climate, water availability, cost drivers, environmental
utilization of the patented Viper Filtration technology on gold requirements, and social factors. Despite the aforementioned
tailings to overcome this challenge and shares the resultant technological growth, the currently available filtration tech-
full-scale plant design, highlighting the features designed to nology is not able to economically meet the needs of many
overcome cost and scalability deterrents. This technology is a operations and projects that would otherwise adopt the tech-
novel mechanical process which complements the vacuum nology. Currently, for tailings to reach the desired moisture
pressure in dewatering the filter cake as it travels along the content for dry stacking, the majority of processes require
belt filter. pressure filtration, with significant upfront and ongoing costs
associated with maintenance. In line with this, the 2017 Study
of Tailings Management Technologies report commissioned
by the Mine Environment Neutral Drainage (MEND)
* Oliver Whatnall Program Secretariat identified “cost and scalability of filtered
[email protected]
tailings is one of the deterrents for mining companies.” Given
the comparative lower costs associated with low pressure
1
Jord International Pty Ltd, Shortland, NSW, Australia Viper vacuum belt filter, this has the potential to alleviate
2
The University of Newcastle, Callaghan, NSW, Australia dependency on high pressure applications and make efficient
 Mining, Metallurgy & Exploration

dewatering and handling of tailings a more economic and cyanide removal and dewatered on the vacuum belt filter
sustainable solution. prior to being stockpiled as filter cake. To facilitate the
design and supply of the 158 m2 vacuum belt filter, the
1.2 Viper Filtration Technology original specification for paste backfill was to wash and
dewater 131 tph of solids to less than 17%w/w moisture
A recently commercialized, patented technology (patent with a P80 ranging from 150 to 190 μm, with 15–20% <
20200166274) utilizes a series of vibration roller assemblies 20 μm and 7–10% < 10 μm.
which are coupled to the top side of a vacuum belt filter. These
rollers, incorporated with other Viper Filtration components, 2.2 Project Justification
complement the vacuum pressure in dewatering the filter cake
as it travels along the belt filter. Given the cost and scalability The tailings deslime and filtration circuit was commissioned
deterrents frequently encountered with pressure filtration of in late 2017, meeting design expectations for feed to the un-
tailings, this technology has the potential to overcome these derground paste fill plant. Since 2017, the specification for
issues, by allowing Viper vacuum belt filter technology to be paste backfill has been revised to a finer PSD, of the order
used in some applications. of 40% < 20 um which cannot be effectively dewatered at the
required rate from the existing vacuum belt filter. Modifying
1.3 Gold Mine and Processing Plant the filter plant in order to achieve this paste specification at
existing design throughput rates is the objective of the project.
This paper presents a case study for an underground gold
mine. The mine produces 500,000 oz (14,174 kg equivalent)
gold per annum via a conventional carbon-in-pulp (CIP) pro-
cessing plant. This processing plant produces slurry tailings of 3 Pilot Plant Testing
which approximately 100 tph is filtered to a low moisture
content and then trucked to the mine to re-pulp as paste 3.1 Objective
backfill.
In 2017, a conventional vacuum belt filter with a filtration The objective of the pilot program was to rigorously test,
area of 158 m2 was installed to wash (i.e., remove cyanide) measure, and record any process performance improvements
and dewater the tailings for paste backfill. Since that time, the achievable using the Viper Filtration technology compared to
specification for paste backfill has been for a finer particle size the pilot-scale conventional vacuum belt filter. The pilot-scale
distribution (PSD) of solids, which could not be effectively testing begins with the control to obtain the baseline before the
dewatered at the required rate from the existing vacuum belt technology is engaged.
filter. Additionally, the pilot testing allows for measurement of
operating and design data, which will form the basis of the
1.4 Pilot Plant Testing full-scale Viper modules and retrofit design. The testing
program considered two sample types, namely, CIP tails
This project commenced with a pilot testing program, which (P80 ~ 40 μm) and tailings thickener underflow (TUF), a
successfully met the objective to rigorously test, measure, and combination of CIP tails and slimes from the existing tail-
record any performance improvements achieved when engag- ings deslime circuit (P 80 ~ 30 μm), both of which are
ing the Viper technology compared to the conventional vacu- deemed acceptable for feeding into the paste backfill plant.
um belt filter. This pilot testing facilitated measurement of PSD information is shown in Fig. 1.
operating and design data, which forms the basis of the full- Specifically, the target process performance is consistent
scale system design and resultant equipment supply of three with achieving < 20% moisture content by weight (%w/w)
vibration roller assemblies for retro-fitting on the existing vac- of the tailings filter cake, required to facilitate trucking, while
uum belt filter. maximizing the throughput rate.

3.2 Equipment
2 Process Summary
The pilot plant is installed and operated at The University of
2.1 Original Design Basis Newcastle in New South Wales, Australia. The base filter is a
vacuum belt filter with 1.6 m2 of filtration area (0.4 m width ×
A portion of the slurry tailings from the process plant are 4.0 m length). Installed in series on the top side of the machine
deslimed and thickened to meet the desired specification are three vibration roller assemblies, each bolted to the filter
for paste backfill. The tailings are washed for residual frame as shown in Fig. 2. The filtration plant also includes the
Mining, Metallurgy & Exploration

Fig. 1 PSD charts comparing CIP

and TUF tailings

filter ancillaries—filtrate receiver vessel, filtrate pump, and slurry, the cake thickness is set by the speed of the belt
liquid ring vacuum pump. and the flowrate of the feed pump. The cake thickness was
a maximum of 12 mm (mm) and was consistently operat-
ing between 9 and 12 mm.
3.3 Process Description ii. Form—the prepared slurry continues to be subjected to
vacuum and results in filtrate being drawn from the slurry.
The slurry tailings are pumped onto the filter at 40%w/w The slurry is formed into a wet “cake”, with or without
solids via the feedbox. The filter feed is continuously present- surface water prior to the vibration roller assemblies.
ed at the feed end of the machine and begins traveling along iii. Mechanical dewatering—the cake travels under a series
the length of the filter via the moving filter cloth and carrier of three vibration rollers which provide consolidation,
belt (see Fig. 3). compaction, and vibration energy into the cake while
Note: this feed solid content is lower than the design basis under constant vacuum pressure. Each vibration roller
of 55%w/w solids, and this testing does not consider the cake assembly can be optimized by changing the roller weight,
washing step. Parallel bench scale vacuum belt filter testwork
is used to inform the full-scale performance expectations and
full-scale equipment design.
The pilot filter has distinct zones/areas:

i. Feed—the feed slurry is spread evenly across the width of

the belt, by the overflowing weir (see Fig. 3), and is im-
mediately exposed to the vacuum. For the given feed

Fig. 2 Pilot vibration roller assemblies Fig. 3 Filter feedbox and overflowing weir distribution system
 Mining, Metallurgy & Exploration

Fig. 4 Filter cake discharge

diameter, surface profile, vibrator type, and vibration in- 4 Results

tensity as well as flexibility to reposition the roller along
the length of the filter. Post the third and final vibration Results from the pilot program were measured and recorded
roller, vacuum continues to be applied to the cake to through the course of processing CIP tailings and TUF tail-
allow further drying. This is the final stage of the filtra- ings. Visual inspection recorded cleaner and more efficient
tion process. cake discharge when operating the pilot plant with the vibra-
iv. Cake discharge—following filtration, the filter cloth fol- tion roller assemblies engaged compared to the conventional
lows an alternate path and separates from the carrier belt. vacuum belt filter. Figure 5 shows the snapshot of the results
The filter cloth travels over the cracking roller and dis- for both sample types, with and without the vibration roller
charges the cake continuously (see Fig. 4). assemblies.
v. Filtrate and vacuum system—all filtrate drawn from Across both tailings types tested, cake moistures > 20%w/
the slurry via the vacuum box is captured in the filtrate w were recorded when operating the pilot plant as a conven-
receiver vessels and is continuously pumped from the tional vacuum belt filter. When operating the pilot plant with
vessels by the filtrate pumps. Vacuum is provided by three vibration roller assemblies engaged, cake moisture con-
the liquid ring vacuum pump, which connects via the tents < 20%w/w were measured for the CIP tailings for the
filtrate receivers through the vacuum hoses and vacu- range of cycle times between 142 and 119 s and for the TUF
um box. The design vacuum suction pressure was − 70 tailings with a cycle time of 119 s.
kPa. When processing CIP tailings, on average, a 4.2%w/w
vi. Filter cloth wash—following cake discharge, the cloth is moisture content reduction was recorded when operating the
cleaned with water by spray bars to ensure process and pilot plant with the Viper Filtration system engaged compared
cloth tracking performance. to the conventional vacuum belt filter. When processing TUF
vii. Filter cloth and belt return—both the filter cloth and tailings, on average, a 5.7%w/w moisture content reduction
carrier belt travel independently along the underside of was recorded when operating the pilot plant with the Viper
the machine. Upon return to the feed end, both travel Filtration system engaged compared to the conventional vac-
under the dam roller, which creates a seal prior to the uum belt filter. Further work is required to understand the why
feed preparation zone the performance difference between CIP tailings and TUF

Fig. 5 Cake moisture versus

cycle time with addition of
vibration roller assemblies
Mining, Metallurgy & Exploration

Fig. 6 Throughput increase

versus cycle time with addition of
vibration roller assemblies

tailings exists to such an extent given the similarity of their A summary of raw results is provided below in Table 1.
respective PSD.
As demonstrated, the cake moisture content increased with
increased belt speed (shown in Fig. 5 as cycle time); however, 5 Full-Scale Design
the gradient of the moisture increase versus belt speed trend is
relatively flat, particularly for the CIP tailings (0.6%w/w The pilot testing program facilitated measurement of operat-
moisture increase with 53 s reduction in cycle time). This ing and design data, which forms the basis of a full-scale
indicates the lack of time dependence for dewatering via the system design and resultant equipment supply for three vibra-
vibration roller assemblies, which is not the case for reducing tion roller assemblies to be retro-fitted on the existing vacuum
cake moisture via conventional “drying.” This is further evi- belt filter. The 3D model screenshot in Fig. 7 illustrates the
denced by the data point in Fig. 5 taken when operating with three green Viper Filtration modules near the discharge end of
only two vibration roller assemblies (rather than three), which the vacuum belt filter. Figure 8 shows the 3D model of a single
had a significantly higher moisture content despite all else Viper Filtration module.
being equivalent. The vibrating action of these rollers improves not only the
This outcome indicates that increased throughput can be moisture content but also the production rate and the discharge
achieved by using the vibration roller assemblies to reduce performance of the cake, particularly for thin cakes. The Viper
the overall filter cycle time. This increased throughput is quan- Filtration technology is also resilient to variance in the feed
tified in Fig. 6. characteristics. The control system will adapt the settings of
Throughput increases in the order of 10–50% compared to the system in response to changes in the process to maintain a
the control were measured. For the CIP tailings sample, these consistent output. Elements of Viper Filtration that deliver
throughput increases were achieved in combination with a these positive outcomes include:
cake moisture content < 20%w/w. For the TUF sample,
throughput increase at the top end of the range resulted in cake i. Substitute the “drying” zone of the filter with more effi-
moisture > 20%w/w (20–22%w/w for increases of ~ 50%). cient mechanical dewatering using the vibration roller

Table 1 Summary of raw test data

Material Cycle time Cake moisture Equipment As tested throughput Washing allowance throughput Moisture reduction vs.
(seconds) (%w/w) configuration increase vs. control (%) increase vs. control (%) control (%w/w)

CIP 142 22.5 Control 0 0 0

142 18.0 3× Viper 15 15 4.5
119 18.4 3× Viper 63 44 4.1
102 18.1 3× Viper 59 43 4.4
89 18.6 3× Viper 62 43 3.9
TUF 142 26.9 Control 0 0 0
142 23.4 3× Viper 12 12 3.5
119 19.7 3× Viper 33 30 7.2
102 20.6 3× Viper 68 50 6.3
89 21.2 3× Viper 70 50 5.7
89 23.1 2× Viper 61 43 3.8
 Mining, Metallurgy & Exploration

Fig. 7 3D model of the Viper

Filtration modules installed on the
vacuum belt filter

assemblies. Less filter area is required to “dry” the filter cake, the CIP tailings, multiplying the above system by eight (i.e., 8
making more of the filter's area available for “forming”. belt filters, each with 3 vibration units) is capable of achieving
ii. Operating with thinner cake optimizes the “form” time, a throughput of 50,000 tpd at ~ 18%w/w moisture content.
increasing throughput with faster belt speed—the thinner Using data from a recent similar project which includes eight
the cake the more efficient the “forming”—the Viper can belt filters with vibration units (see Fig. 9), the operating cost
operate with thin cake as the cake discharge performance of the equipment for consumables including filter cloths is
is optimized by the boundary layer formed at the cake/ approximately $0.10(AUD)/tonne and the installed power re-
cloth interface. quirement was 4 MW.
iii. Maximizing the available vacuum pressure through com-
paction of the interstitial voids further optimizes the
“form”: (a) with more of the filter area used for forming 6 Conclusions
and (b) mechanical “drying” of the cake removing the
porosity of the cake—the vacuum pressure loss via high The testing program in this case study provided the justi-
airflow are minimized. fication to proceed to full-scale site implementation. The
iv. The performance of the mechanical dewatering is shown three Viper Filtration modules have been delivered and
to be more efficient than conventional “drying” deliver- retrofitted to the existing conventional vacuum belt filter.
ing the reduced cake moisture. More broadly, this technology has beneficial industry im-
plications, the most significant being the opportunity to
The technology can be easily scaled up to achieve in- utilize Viper Filtration technology to effectively dewater
creased throughput. For instance, considering dewatering of fine-grained mineral processing tailings at an increased

Fig. 8 3D model of a Viper

Filtration module
Mining, Metallurgy & Exploration

Fig. 9 Layout of 8 off 158m2

vacuum belt filters equipped with
vibratory Viper technology

scale. This technology is one of many to answer the call Compliance with Ethical Standards
from the MEND [1] Program for the industry to improve
technology in support of improved and safer tailings man- Conflict of Interest Oliver Whatnall is an employee of Jord
International Pty Ltd who has a commercial interest in the subject of this
agement. The project identified herein is for improved
manuscript and co-inventor of patent in the subject of this manuscript.
dewatering of tailings for use in paste backfill; however, Kevin Barber is an employee of Jord International Pty Ltd who has a
the technology is also directly translatable for use in commercial interest in the subject of this manuscript and co-inventor of
dewatering tailings for surface disposal or concentrate fil- patent in the subject of this manuscript. Dr. Peter Robinson is a researcher
at The University of Newcastle, who has a commercial interest in the
tration applications.
subject of this manuscript and co-inventor of patent in the subject of this
Large, multiple vacuum belt filter installations are not un- manuscript.
common in the minerals processing industry as they meet the
needs of the modern user, including: Open Access This article is licensed under a Creative Commons
Attribution 4.0 International License, which permits use, sharing, adap-
tation, distribution and reproduction in any medium or format, as long as
i. Simple, scalable technology with more than 25 years of you give appropriate credit to the original author(s) and the source, pro-
operational history in the industry vide a link to the Creative Commons licence, and indicate if changes were
ii. Continuous processing machine with high availability (> made. The images or other third party material in this article are included
in the article's Creative Commons licence, unless indicated otherwise in a
90%) credit line to the material. If material is not included in the article's
iii. Operates with a single filter cloth (typical lifetime 6– Creative Commons licence and your intended use is not permitted by
12 months) statutory regulation or exceeds the permitted use, you will need to obtain
iv. Low supporting infrastructure costs and complexity permission directly from the copyright holder. To view a copy of this
licence, visit http://creativecommons.org/licenses/by/4.0/.
v. Cake discharge from multiple machines can be directed to
an overland conveyor

The technology discussed in this paper delivers these fea-

Reference
tures that can be used to dewater large-scale mineral tailings
potentially suitable for surface disposal (“dry stack”), co-dis- 1. Mine Environment Neutral Drainage (MEND) Program (2017)
posal, or paste backfill as well as concentrate applications. “Study of tailings management technologies.” MEND Report 2501
October 2017. http://mend-nedem.org/wp-content/uploads/2.50.
1Tailings_Management_TechnologiesL.pdf

Publisher’s Note Springer Nature remains neutral with regard to jurisdic-

tional claims in published maps and institutional affiliations.