Sei sulla pagina 1di 67

Technical Report

Saxore Bergbau GmbH


June 2016
Notes to Accompany the Technical Report
This document (“Presentation”) has been prepared by Treliver Minerals Ltd (and associated Group companies), a private limited company incorporated and
registered in England and Wales under company number 07931518 (“Company”) in connection with a presentation to prospective investors. This document
and the information contained in it is solely the responsibility of the Company.
The contents of this Presentation have not been approved by an authorised person within the meaning of the Financial Services and Markets Act 2000
(“FSMA") for the purposes of section 21 of FSMA. Reliance on this Presentation for the purpose of engaging in any investment activity may expose a
prospective investor to a significant risk of losing all of their investment.
Although reasonable care has been taken to ensure the facts stated in this Presentation are accurate and that the assumptions expressed are fair and
reasonable the information in this Presentation, which includes certain information drawn from public sources, does not purpo rt to be comprehensive, and
has not been independently verified and is liable to change.
No representation or warranty, express or implied, is or will be made and to the fullest extent permitted by law, neither the Company nor any of its
representatives, officers, employees, advisers or agents shall accept any liability whatsoever for the accuracy, reliability or completeness of the information or
opinions contained in the Presentation (including any subsequent revisions or amendments) or of any other written information or oral information made or
to be made available to any interested party or its advisers. The information in this Presentation is subject to updating, completion, revision, further
verification and amendment. In providing this information, the Company undertakes no obligation to provide the recipient with access to any additional
information or to update this information or to correct any inaccuracies therein which may become apparent.
This Presentation does not constitute an offer to the public nor requires an approved prospectus under section 85 of FSMA or the Prospectus Rules published
by the Financial Services Authority ("FSA") (as amended) and accordingly this document does not constitute a prospectus for these purposes and has not been
pre-approved by the United Kingdom Listing Authority pursuant to section 85 of FSMA.
This Presentation (including its contents) is only being communicated in the United Kingdom to persons of a kind described in Articles 19(5) (Investment
Professionals), 43 (Members and Creditors of Certain Bodies Corporate), 48(2) (Certified High Net Worth Individuals), 49(2) (High Net Worth Companies,
Unincorporated Associations etc.), 50(1) (Sophisticated Investors), 50A (Self-Certified Sophisticated Investors), 51 (Associations of High Net Worth or
Sophisticated Investors) and any other persons who fall within other applicable exemptions under the Financial Services and Markets Act 2000 (Financial
Promotion) Order 2005 (“Exempt Persons”). It is not intended to be distributed or passed on, directly or indirectly, to any other class of persons. Any
investment activity to which this Presentation relates is available only to Exempt Persons, and other classes of person should not rely or act on this document.
Persons other than Exempt Persons should not take any action nor rely upon this Presentation and should return this Presentation immediately to the
Company.
Statements in this Presentation or otherwise regarding the Company’s or its management’s intentions, beliefs or expectations, or that otherwise speak to
future events, are “forward-looking statements”. Forward-looking statements may be identified by terminology including “could”, “may”, “will”, “should”,
“expect”, “plan”, “except”, “project”, “estimate”, “predict”, “anticipate”, “believes”, “intends” and the negative of these terms or other comparable
terminology. Such statements are based on the Company’s current expectations and involve a number of risks, uncertainties and assumptions and should not
be considered as guarantees of future performance. These statements include, without limitation, statements about the Company’s market opportunity,
growth strategy, competition, expected exploration and development activities, future acquisitions and investments, the adequacy of the Company’s available
cash resources and the availability of future cash resources. Future results and developments discussed in these statements may also be affected by
numerous factors and risks beyond the Company’s control, including political or economic conditions in areas where the Company operates, trade and
regulatory matters, general economic conditions (including commodity prices), and other factors and risks. Readers are cautio ned not to place undue reliance
on these forward-looking statements. The Company does not undertake to update, revise or correct any of the forward-looking information.
A "Certified High Net Worth Individual" is one who has a current certificate of high net worth (signed by his or her employer or accountant) and who has
signed, within the period of twelve months ending on the date of this document, a statement that he or she is of high net worth and understands that he or
she may receive financial promotions which have not been approved for the purposes of section 21 of the FSMA. For these purpo ses, "high net worth" means
having either (i) an annual income of at least £100,000 or (ii) net assets (excluding the value of his or her primary residence and any insurance or pension
rights) of at least £250,000.
A "Sophisticated Investor" is one who has a current certificate signed by an authorised person to the effect that he or she is sufficiently knowledgeable to
understand the risks associated with the type of investment described by this document and who has signed, within the period of twelve months ending on
the date of this document, a statement that he or she is sophisticated in relation to such investments and understands that he or she may receive financial
promotions which have not been approved for the purposes of section 21 of the FSMA.
A "Self-Certified Sophisticated Investor" is one who has signed, within the period of twelve months ending on the date of this document, a statement that he
or she is sophisticated in relation to the type of investment described by this document and understands that he or she may receive financial promotions
which have not been approved for the purposes of section 21 of the FSMA.
This Presentation (including its contents) is confidential and is for distribution only to a limited number of companies or persons and their professional
advisers selected by the Company and its financial adviser for that purpose. It is not intended to be distributed or passed on, directly, or indirectly, to, and
may not be read by, any other persons without the express written consent of the Company. This Presentation is being supplied to you solely for your
information and may not be copied, reproduced, further distributed to any other person or published, in whole or in part, for any purpose.
The distribution of this Presentation in certain non-UK jurisdictions may be restricted by law and therefore persons into whose possession this document
comes should inform themselves about and observe any such restrictions. Any such distribution could result in a violation of the law of such jurisdictions.
Neither this document nor any copy of it may, subject to certain exemptions, be taken or transmitted into Australia, Canada, Japan, South Africa, Singapore, or
the US or distributed to these countries or to any national, citizen or resident thereof or any corporation, partnership or other entity created or organised
under the laws thereof. This Presentation does not constitute or form any part of an offer or invitation to sell or issue or any solicitation of any offer to
purchase or subscribe or otherwise acquire, any ordinary shares in the Company in any jurisdiction
No person other than Company is authorised to give any information or to make any representation other than as contained in this document and, if given or
made, such information or representation must not be relied upon as having been authorised by the Company.
This Presentation has been delivered to interested parties for information only and on the express understanding that they shall use it only for the purpose set
out above. The distribution of this Presentation shall not be deemed to be any form of commitment on the part of the Company to proceed with any
transaction.

By accepting this Presentation, you confirm that you are an Exempt Person and agree to be bound by the foregoing terms and limitations.

The date of this Presentation is 11th November 2014.

2
Executive Summary
Saxore Bergbau GmbH (SBG) was established in Saxony, Germany on 6th December 2011 as a vehicle
to hold tin and multi-metal deposits in Germany. The company is owned 100% by unlisted UK
registered company Treliver Minerals Ltd (TML). TML purchased this company from Australian
registered, unlisted public company Indo Gold Ltd, in November 2013 via a subsidiary Australian
registered company, Saxony Mines Ltd (SML), which has since transferred its holding in SBG 100% to
TML and has been de-registered in Australia. TML is in the process of changing its name to Anglo
Saxony Mining Ltd (ASM).

During May, 2011, applications were made for 3 exploration licences (Erlaubnis). Two of these
(Eibenstock and Kottenheide) were granted in November 2011 and the third (Breitenbrunn) in June
2012. The first two of these licences were valid until 30 June 2016, and the third (Breitenbrunn) is
valid until 31 December 2017, subject to all conditions being satisfied. Collectively these three
licences formed the Westerzgebirge Project. Eibenstock and Kottenheide were relinquished in June
2016 so that effort and funding could be focussed on the main Tellerhäuser Project.

The projects were applied for based on numerous historical workings for tin and associated metals
plus large historical East German (German Democratic Republic - GDR) era resources (not JORC
compliant) totalling approximately 270,000t tin metal reported within the licence areas. Both
greisen and skarn style tin mineralisation is found within the licence areas.

Very large, skarn style, tin-zinc-indium-silver-magnetite deposits were discovered by the GDR-Soviet
joint venture company SDAG Wismut (Wismut) while exploring for and developing a uranium mine
in the area. Subsequent exploration at the Tellerhäuser Prospect by Wismut ultimately resulted in
estimation of 20Mt @ 0.64% Sn (125,000t tin) in skarn and 5.6Mt @ 0.33% Sn (19,000t Sn) in schist
style mineralisation in C1, C2 and Delta Soviet categories, using a 0.15% Sn lower cut-off. The skarn
style mineralisation occurs in three shallowly dipping seams or blankets with a combined strike
length of over 9km, average width of 500m and average thickness of about 3m. It has recently been
recognised by Saxore that there is a separate, later, retrograde overprint of these skarns consisting
of chlorite-amphibole-quartz-cassiterite alteration with significantly higher tin grades that occurs
within structurally controlled pods with an average extent of 200m x 200m x 1-3m. The schist style
mineralisation occurs immediately below the skarn mineralisation and consists of a series of sub-
vertical and/or sub-horizontal quartz-cassiterite+/-tourmaline+/-muscovite+/-chlorite veins in schist.
In addition to tin, the overprinted skarns are host to significant by-product metals including indium,
zinc, silver, iron, gallium and germanium.

Since acquiring the project, Saxore has sourced, analysed and digitised all of the available historical
data which was mostly in the form of paper plans and reports in German and Russian. This digitised
data was then modelled geologically by Beak Consultants based in Freiberg, Germany and
subsequently geochemically block modelled for resource estimation and a report compliant with the
JORC Code 2012 issued by H&S Consultants based in Brisbane, Australia.

The resource report and accompanying Table 1 is available as a separate report but in summary,
resulted in a total resource estimate for Tellerhäuser of 22.1Mt @ 0.46% Sn (101,500t Sn)

3
comprising 19.2Mt @ 0.48% Sn (91,900t Sn) in overprinted skarn and 2.9Mt @ 0.33% Sn (9,600t Sn)
in schist. This is broadly comparable with the GDR era estimates although the grade in the skarn is
lower, partly due to the geostatistical interpolation method used (Kriging) compared with the
manual method used by Wismut. The new estimation reports mineralisation as a mix of Indicated
(6.2Mt @ 0.47% Sn (28,900t Sn)) and Inferred (15.8Mt @ 0.46% Sn (72,600t Sn)), largely based on
drill density and lack of previous QaQc data. However, it is possible that much of the Indicated may
be able to be upgraded to Measured, and some of the Inferred to Indicated, based on recent
channel sample duplicates which show the Wismut tin assays are reliable and may even be
underestimating actual contained tin.

The mineralisation appears to be concentrically zoned such that various higher cut-offs could be
used. At a lower cut-off of 0.5% Sn, the block model resource is 5.9Mt @ 0.92% Sn (54,900t Sn)
which contains 54% of the tin in 27% of the tonnage. Hence as well as being a large scale, low grade
deposit, Tellerhäuser may also be considered as a medium scale, high grade deposit.

From the time of discovery in the late 1960’s until German re-unification in 1990 (when the Soviets
withdrew from East Germany and the project was abandoned) the Wismut joint venture examined
ways to economically extract tin from the skarns. Their final phase of work suggested that a circuit
consisting of preliminary gravity separation followed by cassiterite flotation could produce a 7-10%
tin concentrate, with a recovery of 65-77% Sn and that this could then be upgraded to a saleable
product via fuming.

However, detailed analysis of this work by Saxore has shown that it is seriously flawed, in particular
by not recognising that there are two generations of tin:

 An early mineralisation event associated with prograde garnet-pyroxene-amphibole


skarns that averages around 0.1 to 0.3% Sn (probably largely in the form of silicates and
ultra-fine grained cassiterite) and

 A late mineralisation event associated with retrograde chlorite-amphibole-quartz-


cassiterite that is much higher grade, often plus 1% Sn, which introduced additional tin
as coarse cassiterite.

Because of this, around 0.2% of the tin occurs in the form of silicates and this is fixed no matter what
the actual tin grade is. As the average grade of samples tested by Wismut was around 0.3 to 0.4%
Sn, this means that around half the tin occurs as silicates and consequently that around 50% of the
tin must occurs as cassiterite. Hence, even if all the cassiterite is recovered, recovery should be at
best around 50%. This means that in order to get a 65-77% recovery, about half of the tin in silicates
must be in the concentrate, thus explaining the low concentrate grade. Whenever higher grade
samples were tested by Wismut, results were much better.

Several other problems with the design of the Wismut pilot plant are noted, including poor
comminution circuit design (overgrinding), not separating out magnetite before the gravity circuit,
using cassiterite flotation instead of multi-G gravity separation (as this was not available at the time)
and assuming 0.2% tin in tails is unacceptable (whereas this would be nearly all “unrecoverable tin”
associated with the first stage prograde skarn mineralisation and hence of little economic interest).

4
In fact, the gravity part of the circuit (one spiral and three shaking tables) did produce significant
recoveries of tin to moderate to high grade concentrates and an early report from 1971 suggests
that for skarns with low magnetite and sphalerite content, a 25-30% tin concentrate with a recovery
of 75-81% Sn can be produced from a simple coarse crush followed by dense media separation.

New testwork is currently being conducted by Saxore and its research partners, with encouraging
early results, and it is expected that reasonable recoveries can be achieved once the tin in silicates is
taken into account.

Potential to significantly increase the tin equivalent grade by extracting valuable by-products
including indium, zinc, gallium, germanium, silver, copper and iron (magnetite) can be seen.

An Inferred zinc resource reportable under JORC 2012 guidelines has been estimated as 18.0Mt @
1.12% Zn. This is partially coincident with the tin resource (see Resource Report for details and JORC
compliance).

As Wismut only assayed occasionally for indium, a conservative default value of 1ppm In was
inserted into the block model data for skarn mineralisation so that a rough estimate of indium
content could be obtained. This resulted in an Inferred mineral resource estimate of 15.9Mt @
127ppm In (2,023,000kg In). This is the largest and highest grade indium resource reportable under
JORC 2012 guidelines known (see Resource Report for details and JORC compliance).

Whilst the metallurgical recovery of indium is currently an unknown factor, preliminary work
completed by Saxore suggests that much of the indium is associated with roquesite (copper-indium-
sulphide) and chalcopyrite (copper-sulphide), with values of up to 6,310ppm In being assayed in
hand-picked chalcopyrite rich mineralisation. Concentrating and extracting chalcopyrite is well
established technology as used in most copper mines throughout the world and hence this is not
considered to be a technological challenge. Extracting the indium from such a concentrate is also
proven technology that is currently being used on a commercial basis to extract indium from scrap
photovoltaic material. Hence there is every reason to expect that the indium can be commercially
extracted. Work on the deportment and extraction of indium is ongoing.

In addition to the main Tellerhäuser mineralisation, potential to significantly increase tonnage can
also be seen, in particular:

 Peripheral to existing known mineralisation (which is currently constrained by underground


access considerations).
 Additional skarns below existing known mineralisation and closer to the mineralising granite
contact.
 Between existing known deposits where limited drilling exists.
 By better definition of “feeder zones” which have been inadequately drilled due to the
previous exploration model assuming that mineralisation is confined largely to sub-
horizontal zones.
 At Antonsthal and Breitenbrunn (generally greater than 200m drill spacing) where significant
tin and tungsten mineralisation has already been identified.

5
Hence significant potential exists for increasing the amount of mineralisation within close proximity
of any milling facility.

Germany is a first world, OECD country with very low political risk, excellent infrastructure and a
skilled (but relatively inexpensive) workforce. It is in a moderate tax jurisdiction (around 30%) and
royalties are negotiable with the state (Free State of Saxony). It is likely that royalties will be nil or
very low for a new mining venture.

The Erzgebirge district has been mined continuously since 1168 and thus there is strong local
support for mining. In addition, the state government has recently announced it intends to make
the region a centre for mining excellence and is actively encouraging mining in the district. In fact
the government has recently underwritten the environmental bonds for a new Fluorite mine.

Tin was added to Germany’s strategic minerals list in mid-2014 and there is strong support from
both Germany and the EU for self-reliance on natural resources, especially those considered
strategic.

The aim of Saxore during the next 18 months is to complete a pre-feasibility study on the
Hämmerlein Seam portion of the Tellerhäuser resource which is both above the current water table
and just below it, making dewatering cheap and simple. The existing infrastructure including over
65,000m of underground development, 141,000m drilling/assaying, an operating train used for
transporting tourists, electrical and pumping/dewatering equipment and a recently upgraded water
treatment plant would cost over £100M to replace and should enable fast tracking of any future
mining venture.

6
Contents
1.0 Introduction ................................................................................................................................ 8
2.0 Germany Country Information ................................................................................................. 10
3.0 Mining Regulations ................................................................................................................... 11
4.0 Location and Infrastructure ...................................................................................................... 12
5.0 Geological Setting ..................................................................................................................... 14
6.0 Regional Mineralisation ............................................................................................................ 16
7.0 Breitenbrunn EL ........................................................................................................................ 19
7.1 Geology, Alteration & Mineralisation ......................................................................................... 19
7.2 Previous Exploration & Mining ................................................................................................... 27
7.3 Mineralisation & Resources ........................................................................................................ 29
7.3.1 Introduction .................................................................................................................................................................................................................... 29

7.3.2 Tellerhäuser Deposit ...................................................................................................................................................................................................... 31

7.3.3 3D Modelling & Resource Estimation ............................................................................................................................................................................. 36

7.3.4 Underground Sampling .................................................................................................................................................................................................. 43

7.4 Metallurgy ................................................................................................................................... 49


7.4.1 Historical Work - General ............................................................................................................................................................................................... 49

7.4.2 Dense Media Separation ................................................................................................................................................................................................ 55

7.4.3 Recent Advances ........................................................................................................................................................................................................... 57

7.4.4 Pyro-metallurgy .............................................................................................................................................................................................................. 57

7.4.5 Other Potential Products ................................................................................................................................................................................................ 58

7.5 Environmental & Socio-Economic Considerations...................................................................... 59


7.6 Exploration Potential .................................................................................................................. 60
8.0 Research .................................................................................................................................... 65

7
1.0 Introduction

Saxore Bergbau GmbH (SBG) was established in Saxony, Germany on 6th December 2011 as a vehicle
to hold tin and multi-metal deposits in Germany. The company is owned 100% by unlisted UK
registered company Treliver Minerals Ltd (TML). TML s in the process of changing its name to Anglo
Saxony Mining Ltd (ASM). There was originally an intermediate Australian registered holding
company, Saxony Mines Ltd (SML), but this has recently transferred its entire holding in SBG to TML
and is currently in the process of being de-registered. Treliver purchased these companies from
Australian registered, unlisted public company Indo Gold Ltd (IGL), in November 2013.

While at the Indaba Mining Conference in 2011, IGL personnel familiar with tin mineralisation
recognised the opportunity to secure significant land-holdings in the historically important tin-
tungsten mining district of the Erzgebirge, in the Free State of Saxony, Germany (previously part of
East Germany or the German Democratic Republic (GDR)) following a meeting with Beak Consultants
GmbH (Beak).

The Erzgebirge is a historical mining district and Erzgebirge literally translates as Ore Mountains.
Mining has been conducted within the region for over 800 years, with the first recorded mining
undertaken around 1168AD. It is likely that alluvial cassiterite (tin ore) was sourced from this area as
long ago as 2500BC, during the Bronze Age, although direct evidence for this has not been confirmed
to date. During the twelfth century, mining initially concentrated on silver with by-product lead and
zinc. It is estimated that over 10,000t silver metal has been produced from the district along with
300,000t lead, 150,000t zinc and unspecified amounts of by-product arsenic, bismuth, indium and
germanium. Tin was discovered during the thirteenth century and it is estimated that around
350,000t tin metal has been produced from the district up until 1990 as well as over 23,000t
tungsten metal. During the East Germany (GDR) era between 1946 and 1989, approximately
250,000t uranium metal was produced from the district, making it one of the premier uranium
mining areas in the world. It can thus be seen that the Erzgebirge is aptly named and is a world class
source of tin, silver, uranium and other metals.

Following recognition of the large metal inventory in the region, IGL applied for three Erlaubnis
(equivalent of exploration licences) in the western part of the Erzgebirge, covering the largest known
deposits and the largest tin geochemical anomalies associated with an abundance of medieval tin
workings. Two of these Erlaubnis - Eibenstock and Kottenheide, which cover the tin geochemical
anomalies with numerous greisen style tin deposits, were granted in November 2011. The third,
Breitenbrunn, covering large skarn deposits, was granted in June 2012 following a dispute resulting
from over-lapping applications by another party, which was resolved by one deposit (Pohla
Globenstein) being excluded from IGL’s original applications.

Saxony Mines Ltd (SML), an unlisted Australian registered company, was formed as a dedicated
Germany focussed tin-tungsten exploration and development vehicle on 2nd August 2011. It was at
that time a 100% owned subsidiary of St Piran Mines Ltd (SPML) which was in turn a 100% subsidiary
of Indo Gold Ltd (IGL). A local unlisted operating company, Saxore Bergbau GmbH (SBG), was
established and registered in Saxony on 6th December 2011.

8
Treliver Minerals Ltd (TML), an unlisted, UK registered company, subsequently purchased SML and
its subsidiary SBG from IGL via SPML on 14th November 2013 and hence SML and SBG are now 100%
owned subsidiaries of TML. As noted above, SML is in the process of being de-registered and TML is
in the process of changing its name to Anglo Saxony Mining Ltd (ASM).

9
2.0 Germany Country Information

The Federal Republic of Germany (Germany) is the most populous member of the European Union
(EU) with approximately 82 million people and covers an area of about 357,000km². The official
language is German.

A region known as Germania was occupied by several Germanic speaking peoples before 100AD.
These tribes expanded southwards and established successor kingdoms throughout Europe.
Beginning in the 10th Century, German territories became part of the Holy Roman Empire. During
the 16th Century, the northern parts of Germany became the centre of the Protestant Reformation
while the southern parts remained largely Roman Catholic. In 1871, most of the German states
formed the German Empire. After the Second World War, in 1945 Germany was split into two
countries – East and West Germany – and was eventually re-unified in 1990.

Germany is the World’s fourth largest economy by nominal GDP and the fifth largest by purchasing
power parity. It is the World’s third largest importer and exporter. It was a founding member of the
European Community in 1957 which became the EU in 1993. It is also a member of the United
Nations, NATO, the G8, the G20, the OECD and the Council of Europe, and took a non-permanent
seat on the UN Security Council for the 2011–2012 term.

Germany is a first world country with high per capita GDP and low levels of corruption.
Infrastructure is excellent and skilled labour abundant. Germany and the EU in general have
recently made a push toward self-sufficiency with regard to primary resources, thus initiating a
renewed interest in exploration and mining activity. Tin has recently been added to the strategic
minerals list in Germany.

Germany is a relatively moderate tax regime, with corporate tax being around 30% depending on
the location of the business. No set mineral royalties currently exist but the state (Free State of
Saxony in this case) may levy royalties and this will need to be negotiated. Advice we have had to
date suggests that royalties will be nil or very low for a start-up mine.

The cost base for labour in Germany is relatively low, with the cost of skilled underground miners
around one third of that in Australia. Thus mining costs should be considerably lower than in
Australia.

The climate of Germany is temperate, with mild summers and cool winters. The Erzgebirge district is
generally covered in snow between November and March.

10
3.0 Mining Regulations

The agency administering mining and exploration in Saxony is the Saxonian Mining Authority
(Sächsische Oberbergamt) which is part of the Saxonian State Ministry for Economic Affairs, Labour
and Transport (Staatsministerium für Wirtschaft, Arbeit und Verkehr). This agency administers the
Federal Mining Law.

Two types of licence exist:

 Erlaubnis – equivalent to an Exploration Licence


 Bewilligung – equivalent to a Mining Lease or Production Licence

Applications for these licences must include a proposed work program which must be adhered to as
a minimum commitment and must include the proposed timing for the work to be undertaken. In
the case of a Mining Lease application, details of the mineralisation must be included along with
proof that the minerals are extractable and a program for such extraction including a reasonable
time frame.

Exploration Licences give the holder sole right to explore for the minerals named and Mining
Licences give the holder the sole right to mine the minerals named.

There is no first in line priority so that if two applications are made for the same area, these will be
decided based on the proposed work program and the financial and technical ability of the
applicants.

MLs take precedence over ELs and an ML may be applied for over ground held as an EL. However, if
someone applies for an ML over another parties EL, that party will be given 3 months in which to
lodge a competing ML application which will then be given priority. Thus tenure under an EL is
reasonably secure as the equivalent of a feasibility study is required for grant of an ML.

ELs are generally granted for up to 5 years depending on the proposed work program and are
renewable for an additional 2 years if all conditions have been met. MLs are generally granted for 50
years.

11
4.0 Location and Infrastructure

The Erzgebirge district of the Free State of Saxony is located in the southern part of the old “East
Germany” or German Democratic Republic (GDR) along the border with the Czech Republic
approximately 230km SSW of Berlin and 110km WNW of Prague (Figure 1).

Figure 1: Location of the Westerzgebirge Project

International airports are located at Dresden, the state capital, and Leipzig, the second city of
Saxony. Both these airports are connected to Freiberg, the location of Saxore’s main office, by train.
Freiberg is a regional centre with a population of around 40,000. It is the headquarters of the state
Mines Authority, is home to the World’s oldest school of mines, has a world class mineral museum
and has commercially available metallurgical and mineral processing test facilities up to pilot scale.

12
The main areas of interest are located approximately 70km WSW of Freiberg and these are generally
accessed by good quality roads ranging from 6 lane autobahns to local sealed 2 lane roads (Figure 2).

20km

Figure 2: Project and Local Office Location

The main mineralisation at Tellerhäuser is currently partly accessible (Hämmerlein seam) by an adit
that extends 3km under a hill and that has an operating train service operated as a tourist attraction.
The adit extends a further 6km past the current tourist area but this is currently blocked off and
flooded. Several internal shafts then access the other mineralisation seams beneath the main adit
level. All up, it is estimated that approximately 60-70km of access, including cross-cuts for drilling
etc, has been excavated for tin exploration. It is estimated that about one million cubic metres of
water is present within these voids (including old uranium stopes) and this would require de-
watering to access the lowest levels of the old mine. Mine seepage water is currently being
collected and treated before release into the local drainage system. This plant was originally
removing heavy metals plus radioactive material but is currently only removing heavy metals (mainly
arsenic) as no radioactivity is being detected. Its current capacity is up to 60m3 per hour for diluted
mine water and about 30m3 for undiluted mine water. The normal outflow is 11-12m3 per hour.

This mine and water treatment infrastructure is owned by Wismut GmbH, a federal government
owned company that is tasked with cleaning up the old uranium mining activity that was conducted
by SDAG Wismut, the joint Soviet/GDR company that explored and mined in the area during the GDR
era. It is held under a mining lease for uranium only, but no uranium mining is allowed under the
company’s current mandate. The currently accessible part of this mining infrastructure is leased to a
local co-operative who operate it as a tourist mine. Wismut has informally offered to sell the
infrastructure to Saxore for one Euro, but as there is a significant environmental liability that would
go with the purchase, this needs to be fully assessed before a purchase is made.

13
Surface infrastructure is excellent with sealed roads and mains power (10kV) in place to the existing
tourist mine. Within 5km of the existing adit portal there is a large pumped-storage hydro power
station at Markersbach connected to the 380kV electricity network of Europe.

5.0 Geological Setting

The area is located within the Saxo-thuringian domain of central Europe in the Hercynian-Variscan
orogen (Figure 3). This orogen hosts most of the tin mineralisation in Europe including the giant
Cornwall tin district.

Figure 3: Tectonic Setting of the Westerzgebirge Project

In more detail, the Erzgebirge consists of a sequence of metamorphosed Palaeozoic (Cambrian to


Ordovician) sediments consisting of shale, siltstone-sandstone and some limestone lenses which
have all been intruded by granitoids of Carboniferous to Permian age (Figure 4). These granitoids
are the source of the mineralising fluids which have deposited the main tin and associated
mineralisation. Later cross-cutting sub-vertical faulting is the main control on the uranium
mineralisation.

14
Figure 4: Simplified Geology of Saxony (from Saxony Mining Authority)

A major orogeny (Hercynian-Variscan Orogeny) deformed the Palaeozoic sedimentary sequence


during the late Devonian to early Carboniferous. This was caused by the collision between
Gondwana and Laurasia during the formation of the supercontinent Pangaea. The early granitoids
(Carboniferous and earlier) are deformed by this orogeny whereas the later, mineralising granitoids
(Permian) are generally undeformed and are classed as post tectonic.

15
6.0 Regional Mineralisation

Saxony is one of the oldest continually mined regions in the world, with the first recorded mining
taking place in 1168. There is some evidence that tin was sourced from this district during the
Bronze Age around 2500BC but there is still controversy surrounding these claims.

The main metallic mineralisation in Saxony is found in the Erzgebirge district, which literally
translates as “Ore Mountains”. The summits of these mountains define the approximate location of
the Germany - Czech Republic border and the range is also known as the “Ore Mountains” or Krušné
Hory in the Czech Republic itself.

While records are incomplete, reported production from the area includes:

 Uranium >250,000t
 Silver >10,000t
 Lead >300,000t
 Zinc >150,000t
 Tin >300,000t
 Tungsten >23,000t

Current resources (from Saxonian Ministry of Economy and Labour, not reportable under JORC or
NI43-101 guidelines) are estimated to include:

 Uranium >33,000t
 Silver >2,700t
 Lead >392,000t
 Zinc >736,000t
 Copper >236,000t
 Tin >773,000t
 Tungsten >229,000t
 Indium >287t
 Gallium >7t
 Lithium >118,000t
 Fluorite >8,500,000t
 Barite >10,700,000t

The majority of the mineralisation is associated with the large Permian batholiths and is of magmatic
hydrothermal origin. Three main batholiths are recognised within the district, associated with the
West, Central and East Erzgebirge mineralisation districts. Figure 5 shows the distribution of known
tin mineralisation within the Saxon Erzgebirge (data from Saxonian State Ministry for Economic
Affairs, Labour and Transport) and clearly defines these three main areas of mineralisation.

The batholith associated with the West Erzgebirge is the best exposed, due to a generally deeper
level of erosion in the west. The East Erzgebirge batholith is at an intermediate level of erosion with
more sub-volcanic rocks exposed at surface and the Central Erzgebirge batholith is the least eroded

16
and only limited outcrops occur. This is reflected in a generally lower level of tin anomalism at
surface (Figure 6, data from Saxonian State Ministry for Economic Affairs, Labour and Transport).

Figure 5: Distribution of Tin Mineralisation in the Saxon Erzgebirge

Figure 6: Regional Tin Stream Sediment Geochemistry in the West Erzgebirge


(Background Image) showing Licence Locations and Known Tin Mineralisation

17
Saxore has secured a large part of the known West Erzgebirge tin mineralisation and associated tin
geochemical anomalism as shown on Figures 5 & 6 and Table 1. All up, more than 270,000t tin metal
has been estimated to occur within the three licences in skarn, greisen and alluvial type deposits.
However, the majority of this is located within the Breitenbrunn licence which has been the focus of
most exploration to date. It has recently been decided to relinquish the two western licences in
order to concentrate effort on the main Tellerhäuser mineralisation.

Table 1: GDR Era Estimates of Mineralisation within the Westerzgebirge Project Area

18
7.0 Breitenbrunn EL

The Breitenbrunn EL has an area of 71.4km² and was granted on 4th June 2012. The licence is valid
until 31st December, 2017 so long as all obligations are met.

This licence was applied for based on the large metal inventory identified by previous explorers
(mainly by SDAG Wismut – a Soviet/GDR joint venture). This inventory totals over 243,000t tin
metal, 23,000t tungsten metal, 2.8Mt magnetite, 91,000t zinc plus large but undefined inventories
of indium, gallium, germanium, silver, lead, copper, fluorite and barite.

Although these estimates are not reportable under JORC or NI43-101 guidelines, the previous work
was generally conducted to a high standard, e.g. drilling to 25m x 12m in places. Saxore has sourced,
compiled and digitised all of the old information and has used this to obtain a JORC compliant
resource as discussed below.

7.1 Geology, Alteration & Mineralisation

The geology of the Breitenbrunn EL (Figure7) consists of a sequence of flat lying to shallowly dipping
metamorphosed pelitic sediments, with limestone interbeds, of Cambrian to Ordovician age. The
original sediments have been metamorphosed to schists and marble/calc-silicate/gneiss respectively
and have been intruded by Permian monzonitic to syenitic granitoids. The calc-silicate/gneiss has
been partially metasomatised to skarns in proximity to the fertile granitic intrusions.

Figure 7: Breitenbrunn EL – Simplified Geology (Base map from Geological Survey of Saxony)

19
The majority of the known tin mineralisation in the licence is hosted by skarns. These occur within
generally sub-horizontal to shallowly dipping calc-silicate/gneiss units (with minor schist interbeds)
that have been preferentially metasomatised by the mineralising fluids as they were released from
the still cooling granite. The skarn occurs as lenses or irregular “alteration fronts” within the calc-
silicate units rather than as a single well defined unit (e.g. Figures 8 and 9) and it is difficult to
separate out individual skarn bands. Hence the entire package of calc-silicate/gneiss, skarn and
minor schist is referred to as a skarn package and modelled as such. As the actual skarn lenses are
dark coloured and have a high density compared to the calc-silicate and gneiss (which are light
coloured and have a low density (Figures 8 and 9)), it is likely that the package can be mined as a
single unit and that the skarn can then be separated out by simple dense media, photometric or XRF
sorting, possibly while still underground (see later discussion in mineral processing section).

Figure 8: Skarn Package (>5m thick) showing Lenses of Skarn (dark) within
Calc-silicate Host Rock (light), Hämmerlein Seam, Tellerhäuser Project

20
Figure 9: Skarn Alteration along Fractures in Calc -silicate/Gneiss, Hämmerlein Seam, Tellerhäuser
Project (note early calc-silicate alteration (pink) followed by later amphibole alteration (black))

The skarn mineralisation is complex, with several different types of skarn being present depending
on the original lithology of the limestone, distance from granite, temperature of the mineralising
fluids and different Eh-pH conditions etc. In broad terms, the skarn package can be thought of as
having four distinct events:

1. An early calc-silicate alteration (the pink alteration in Fig.9) which is not associated with any
significant mineralisation.
2. The main prograde garnet-pyroxene-amphibole(hornblende) stage that is weakly tin
mineralised (0.1% to 0.3% Sn). Tin is often present in silicate form (e.g. malayaite, stokesite)
and in the lattice of or as fine inclusions within other silicate minerals (e.g. amphibole,
epidote, titanite and rutile). It is also present as fine cassiterite crystals (10-100 microns).
3. The introduction of magnetite-sphalerite-cassiterite plus or minus other sulphides (Cu, As,
Fe) in varying proportions. Tin occurs as both coarse discrete cassiterite crystals and
sometimes as ultrafine cassiterite inclusions within magnetite and other iron oxides.
4. A late, retrograde overprinting event which is comprised largely of chlorite-
amphibole(tremolite?)-quartz-cassiterite. Tin appears to occur largely in the form of very
coarse grained cassiterite crystals up to several cm across (e.g. Figures 16 & 17). This
normally has high grade tin mineralisation (plus 0.5% Sn, often plus 1-2% Sn) and forms the
majority of the potentially mineable deposit.

It is unclear whether the magnetite-sphalerite mineralising event is earlier than or contemporaneous


with the late retrograde chloritic overprint. It definitely cross-cuts the prograde skarn and hence
must be later than that event. Examples of the various types of skarn and mineralisation are shown
as Figures 10 to 16.

21
Pyroxene (Hedenbergite/Diopside)

Amphibole (Hornblende)

Garnet (Andradite)

Figure 1 0: Garnet-Pyroxene -Amphibole Skarn

Pyroxene (Hedenbergite/Diopside)

Quartz-Chlorite+/-Cassiterite Overprint

Figure 11: Pyroxene Skarn with Quartz-Chlorite+/-Cassiterite Overprint

22
Figure 12: Magnetite Skarn

Sphalerite

Magnetite

Figure 13: Sphalerite-Magnetite Skarn

23
Figure 14: Sulphide-Amphibole-?Chlorite-Magnetite Skarn (with Chalcopyrite -Roquesite)

Calc-Silicate/Skarn
Coarse Cassiterite

Quartz-Chlorite+/-Cassiterite Overprint

Figure 15: Quartz-Chlorite-Cassiterite Retrograde Skarn Alteration Front


Overprinting Calc-Silicate /Prograde Skarn, Hämmerlein Seam, Tellerhäuser Project

24
Figure 16: Coarse Cassiterite in Quartz Veining with Chlorite alteration
Overprinting Prograde Skarn, Hämmerlein Seam, Tellerhäuser Project

The fluids responsible for the skarn mineralisation (both prograde and retrograde) require a
plumbing system to travel along, and this may be in the form of faults, breccia pipes and/or other
structural discontinuities. These structures are often mineralised themselves by the fluids passing
through them, and they can often be high grade due to repeated pulses of fluid flow. In addition,
the skarn beds can often act as aquitards, creating a trap for the fluids which then pond in the
underlying rock sequence.

These plumbing systems or “feeder zones” have been identified in the vicinity of the skarn
mineralisation, in particular associated with the Hämmerlein seam, and they vary from sub-vertical
to sub-horizontal in nature (Figures 17 to 19). As most of the drilling was designed to test the sub-
horizontal skarns, it is often not optimal for testing the sub-vertical component. Hence, while it is
recorded that around 13% of the tin mineralisation occurs as “schist ore” from the feeder zones, it is
possible that much more of this style of mineralisation exists but has not been adequately tested by
the drilling to date.

The “schist ore” or feeder zone mineralisation is much simpler than the skarn mineralisation and
consists of sub-vertical and sub-horizontal veins and disseminations of quartz+/-feldspar+/-
tourmaline+/-mica with reasonably coarse cassiterite within mica schist (metamorphosed pelitic
sediments). In the backs (roof) of the existing underground workings, sheeted quartz-tourmaline-
cassiterite veins can be seen within schist (Figure 17) while in the wall of the main adit, sub-
horizontal quartz-cassiterite veins can be seen (Figure 18). Some samples in the Wismut offices
show very coarse cassiterite-quartz-feldspar veins from the “schist ore” (Figure 19). These represent

25
good targets for additional mineralisation with potentially high grades and simple metallurgical
characteristics.

Quartz-Tourmaline-Cassiterite Veins

Figure 17: Sub-vertical Sheeted Quartz-Tourmaline-Cassiterite Veins


- Backs (Adit Roof), Hämmerlein Schist Mineralisation

Cassiterite

Figure 18: Sub-horizontal Quartz -Cassiterite Veins,


Hämmerlein Schist Mineralisation

26
Figure 19: Quartz-Feldspar-Coarse Cassiterite
Vein from the Hämmerlein S chist

7.2 Previous Exploration & Mining

SDAG Wismut (an East German – Soviet joint venture company formed for uranium mining and
exploration) began exploring for uranium during the late 1940s and mined several small veins in the
district containing a few tonnes of uranium. A shaft was sunk at Pöhla-Globenstein but found only
minor uranium mineralisation. However, a magnetite-skarn was discovered and the shaft was
transferred to an East German iron company for iron ore exploration.

During the 1960s, drilling identified radioactive anomalies in the Hämmerlein and Tellerhäuser areas,
to the south of Pöhla-Globenstein. It was decided to explore these deposits underground using an
adit which was started in 1967. Approximately 3km from the portal, the adit intersected minor
uranium mineralisation at Hämmerlein which was mined out during the exploration phase resulting
in production of about 15 tonnes of uranium. However, significant tin mineralisation was discovered
in skarns and schists in the same area. These skarns were also found to host significant quantities of
zinc, magnetite, indium and cadmium. This mineralisation was intensively explored from
underground during the 1970s and 1980s and plans were made to set up a major tin mine.
However, although the total resource base discovered was very large and the grades better than
other producing tin mines in East Germany, mineralogical and metallurgical testwork concluded that
only a low grade tin concentrate could be produced, even though recoveries were good (see later
discussion on mineral processing).

From Hämmerlein, the main adit was driven further during the 1970s to investigate the uranium
mineralisation at Tellerhäuser, which was intersected at about 7.5km from the portal. This

27
mineralisation was much larger than at Hämmerlein and two underground shafts were sunk to
access deeper parts of the deposit (~500m below the adit level). Mining began in 1983 and 1,200
tonnes of uranium were produced until the end of production in 1990. It was estimated that over
5000t uranium remained at Tellerhäuser in 1990. Similar to Hämmerlein, significant tin and
magnetite mineralisation was discovered in skarns at Tellerhäuser. Some of the magnetite was
mined as additive to concrete for nuclear power plants constructed by the East German Government
in the 1980s. Small amounts of silver were also extracted, but high arsenic contents of the silver ore
made processing very expensive and production of silver was stopped in the late 1980s. An ore
shoot with massive native arsenic and silver was discovered in 1990, shortly before the end of
uranium mining. Some of this material was mined, but was never processed.

After production ceased, the area was rehabilitated, including the flooding of the mine, removal of
the mine buildings and covering of the waste rock area. A water treatment plant was set up to
remove uranium, radium and arsenic from the mine water. Because of the low water flow from the
mine, it was possible to replace the initial active chemical water treatment plant by a passive
biological unit (wetland). However, this plant did not perform to expectations and hence an active
chemical treatment plant has recently been re-established.

28
7.3 Mineralisation & Resources

7.3.1 Introduction

Known mineralisation and resources within the Breitenbrunn licence are associated with two
regional packages of schist and interbedded skarn shown as Skarn Package 1 and Skarn Package 2 on
the plan and section below (Figures 20 & 21). These units have been folded into an open domal
structure with shallow to moderate dips to the south and west within Saxore’s licence area. The
mineralising Eibenstock Granite is interpreted to underlie most of the area at depths of between
zero and 1000m. Within the skarn packages, individual skarn beds between a few cm and several
10’s of metres occur within a package of dominantly pelitic sediments (now metamorphosed to
schist). These beds are interpreted as originally being limestone interbeds. More details of the
detailed geology, alteration and mineralisation may be found in Section 7.1 above.

Three main deposits, Tellerhäuser, Breitenbrunn and Antonsthal, were discovered and explored in
GDR times. These are all associated with Skarn Package 1. In addition, the Pöhla Globenstein
deposit, which is currently excised from Saxore’s ground, was also discovered and evaluated. That
deposit is located within Skarn Package 2 which is stratigraphically below Skarn Package 1. Limited
exploration has been undertaken between the known deposits and it is considered highly likely that
additional mineralisation will be found with further exploration (see later section on exploration
potential).

Figure 20: Surface Projection of Skarn Mineralisation and Main Adit on Local Geology Plan, Breitenbrunn EL

29
Figure 21: Longitudinal Section A-B (see Fig 21 for location) Showing 2 Main Skarn Packages and Known Mineralisation

The historical, non-JORC estimates for mineralisation within the Breitenbrunn EL are presented
below in Table 2 (see Figures 20 and 22 for locations).

Tonnes Grade Tonnes Tonnes Grade Tonnes


Prospect Category
(Mt) Sn (%) Sn (Mt) W (%) W

Antonsthal 14.7 0.19 28,000 D


6.2 0.37 23,000 D
Breitenbrunn 22.3 0.25 55,000 D
Hämmerlein 7.3 0.38 27,802 C1
5.0 0.47 23,305 C2
1.5 0.25 3,803 D
2.1 0.15 3,118 c1+c2
15.9 0.37 58,028 Sub-total
Tellerhäuser 1.9 0.84 15,889 C1
8.4 0.64 53,746 C2
4.4 0.57 24,939 D
4.4 0.17 7,519 c1+c2
19.1 0.53 102,093 Sub-total

TOTAL 72.0 0.34 243,121 6.2 0.37 23,000


Table 2 – Total Historical Mineralisation Estimated during the East German Era within the current Breitenbrunn Licence

The East German categories, while not directly equivalent to JORC or NI43-101 categories, can be
considered as broadly similar to:

 C1: measured plus indicated


 C2: indicated plus inferred
 D: inferred plus exploration target
 c1 and c2: sub-grade and therefore of little interest.

Drilling at Breitenbrunn and Antonsthal is very broad spaced and the resources are considered as
exploration targets only. However, as the tonnage decreases and grade improves with additional
drilling at Tellerhäuser, the same is likely at these other two deposits and hence potential for higher
grade satellite feed is considered to be high (see later discussion under exploration potential).

Due to the confusing naming conventions used in the past for different parts of the mine and for
both uranium and tin mineralisation, a simplified convention has been adopted by Saxore.

30
The tin deposit previously referred to as Hämmerlein-Tellerhäuser is now called simply the
Tellerhäuser project or deposit, the three skarn seams are named the Hämmerlein, Zweibach and
Dreiberg seams or skarns (these are shown on the plan and cross-section in Figures 22 and 23) and
the schist mineralisation at Hämmerlein is referred to as the Hämmerlein schist. The Breitenbrunn
and Antonsthal deposit’s names remain unchanged.

7.3.2 Tellerhäuser Deposit

The Tellerhäuser deposit consists of three shallowly dipping (0° to 20°, average 10°) units comprising
dominantly calc-silicates and gneiss (with minor schist interbeds), the majority of which has been
altered to skarn (see photographs in Figures 8-16 above and the discussion in section 7.1). These
units have been named the Hämmerlein, Dreiberg and Zweibach seams (Figures 22 & 23). They are
each between 1m and 11m thick (average 3.3m), 300-800m wide and 2000m-5000m long. In
addition, a zone of schist hosted mineralisation (feeder zone?) is located immediately below the
Hämmerlein seam. Higher grade parts of these seams are about 200m x 200m in extent and are
typically seen as retrograde chlorite-amphibole-quartz-cassiterite overprints of the prograde garnet-
pyroxene-amphibole skarns.

The tin mineralisation at Tellerhäuser has been extensively drilled from both surface and
underground, down to 25m x 12.5m spacing in some areas. In total, the following statistics have
been extracted from the database:

 65km underground access


 141,342m drilling in 2,112 bore-holes
 3,083m channel sampling in 1,326 channels
 60,085 samples assayed
 151,916 individual assays

In addition, some limited trial mining was undertaken in the Hämmerlein skarn (49,190t @ 0.31% Sn)
and schist (15,900t @ 0.34% Sn) as a test of preferred mining techniques as well as to obtain a bulk
sample for trial processing.

A plus 9km adit and substantial underground access is in place (Figures 23 & 24) but this has been
partially flooded and is now only accessible into part of the Hämmerlein seam, which is currently
kept open as a tourist mine.

The total replacement cost of this work is estimated at over £100M.

The underground access and mine infrastructure is still in place and in working order in the upper
levels of the mine, including the 3km production size adit, an operating train service (currently used
to transport tourists), much of the electrical and de-watering infrastructure and a recently upgraded
water treatment plant. This could all potentially be used to fast track a small to medium scale
mining operation.

31
Figure 22: Simplified Geology, Tellerhäuser Deposit, Showing the Three Main Skarn Horizons

Figure 23: Simplified Longitudinal Section through the Tellerhäuser Deposit, Showing the Three Main Skarn Horizons
Dipping Shallowly to the South

32
Table 3 shows the historical estimates for the mineralisation at Tellerhäuser broken down into the
three seams and excluding the sub-grade mineralisation (c1 and c2) included in table 2. It can be
seen that the total skarn mineralisation in the three seams is 19.7Mt @ 0.64% Sn (125,000t tin) plus
an additional 5.6Mt @ 0.33% Sn (19,000t tin) in schist associated with the Hämmerlein seam. The
deep Dreiberg seam is the highest grade part of the deposit, possible due to proximity to the source
granite (Figure 24), and is estimated to contain 8.0Mt @ 0.88% Sn (71,000t tin).

Table 3: Summary of GDR Era Tellerhäuser Mineralisation Estimates by Seam (C1, C2& D categories only)

While estimations of other metals are not available for these deposits, summary cross-sections
showing metal distribution (e.g. Figures 24-28) and a review of the digital database show that the
following grades occur:

 Magnetite up to 60%
 Zinc up to 20%
 Indium up to 2,000g/t
 Gallium up to 50g/t
 Germanium up to 20g/t
 Silver up to 150g/t
 Copper up to 0.2%

33
Figure 24 – Distribution of C1, C2 and D Mineralisation at Tellerhäuser showing Location of Cross Sections

Figure 25 – Cross Section A-A’, Hämmerlein Seam, Tellerhäuser Deposit

34
Figure 26 – Cross Section B-B’, Hämmerlein Seam, Tellerhäuser Deposit

Figure 27 – Cross Section C-C’, Dreiberg Seam, Tellerhäuser Deposit

35
Figure 28 – Cross Section D-D’, Dreiberg Seam, Tellerhäuser Deposit

7.3.3 3D Modelling & Resource Estimation

All data for Tellerhäuser has now been digitised from paper sections, plans and notebooks/assay
sheets and has been incorporated into a 3D model.

The screen shots below show the extent of the underground workings and drill information.

Figure 29 - Extent of Drilling and underground Excavations, Tellerhäuser Project

36
Figure 30 – Oblique View of Underground Development, Tellerhäuser, Note 2 x 500m Internal Shafts

1km

Figure 31 – Close-up of Hämmerlein Seam C1 area, Tellerhäuser, Showing Extent of Development (Blue), Drilling (Green) and
Trial Mining (Red)

37
Following the digitisation, a broad geological model was constructed (i.e. skarn “packages” wire
framed) by Beak Consultants in Freiberg and geostatistical block modelling and estimation of mineral
resources undertaken by H&S Consultants (H&S) in Brisbane, Australia. The mineral resource
estimates (MRE) have been reported under JORC 2012 guidelines and these are available as a
separate report and press release.

In summary, the MRE for the Tellerhäuser deposit as a whole is:

Category Mt Tin % Tin tonnes Density (t/m3)


Indicated 6.2 0.47 28,900 2.95
Inferred 15.8 0.46 72,600 3.02
Total 22.1 0.46 101,500 3.00
Table 4 - Tellerhäuser Mineral Resource Estimate

The grade-tonnage curves for the total resource is shown below in both tabular (Table 5) and
graphical (Figure 32) formats:

Table 5 - Tellerhäuser MRE Grade Tonnage Estimates

Figure 32 - Tellerhäuser Grade-Tonnage Curve

It can be seen that at the same lower cut-off as Wismut (0.15% Sn), the tonnage is higher and the
grade lower than the Wismut estimates (Saxore: 30.4Mt @ 0.38% Sn vs Wismut: 25.3Mt @ 0.57%
Sn). This is partly due to the wireframes of Saxore being geological rather than grade constrained
(hence including some un-mineralised skarn) and partly due to the geostatistical (Ordinary Kriging vs
manual) geochemical modelling techniques of H&S and Wismut respectively.

38
The grade is actually concentrically zoned when looked at in a spatial view, with high grade (plus
0.5% Sn blocks showing a distinct pattern of approximately 200m x 200m pods, probably structurally
controlled, in both the Hämmerlein and Dreiberg seams (Figures 33 & 34).

Block Tin Grade

Fig 33 - Block Model for Hämmerlein Seam, Showing Block Grade

39
Block Tin Grade

Fig 34 - Block Model for Dreiberg Seam, Showing Block Grade

Thus it should be possible to mine these higher grade pods selectively. The 0.5% cut-off resource is
5.9Mt @ 0.92% Sn (54,900t tin), which contains over half the tin in about a quarter of the tonnage
compared with the 0.2% Sn cut-off resource. As previously noted, it appears that the prograde
garnet-pyroxene-amphibole skarn is grading around 0.2% Sn (generally between 0.1% and 0.3%) and
that most of the tin in this unit occurs either within silicates or as ultra-fine grained cassiterite. It
also appears that the retrograde chlorite-amphibole-quartz-cassiterite overprint is much higher
grade, with tin occurring generally as coarse grained cassiterite. Hence it is reasonable to assume
that much of the tin in the below 0.5% Sn blocks in skarn will be “un-recoverable” and hence
focussing on the higher grade blocks will not result in losing much of the “recoverable” tin.

40
However, this is not the case in the schist hosted mineralisation where most of the tin occurs as
coarse cassiterite.

As such, it is possible to consider a lower tin cut-off grade for the schist mineralisation compared
with the skarns and hence Table 6 below shows the estimated resources (indicated plus inferred) for
each skarn seam assuming a 0.5% Sn cut-off and for the Hämmerlein schist mineralisation assuming
a 0.3% Sn cut-off:

Seam Indicated Inferred Total


Tonnes Grade Tin (t) Tonnes Grade Tin (t) Tonnes Grade Tin (t)
(million) (% Sn) (million) (% Sn) (million) (% Sn)
Hämmerlein 0.75 0.85 6,400 0.75 0.81 6,100 1.50 0.83 12,500
Skarn
Hämmerlein 0.97 0.48 4,700 0.16 0.56 900 1.13 0.49 5,500
Schist
Hämmerlein 1.72 0.64 11,000 0.91 0.77 7,000 2.63 0.68 18,000
Total
Dreiberg 0.65 1.11 7,200 3.17 0.97 30,800 3.82 0.99 38,000
Skarn
Zweibach 0.08 0.53 40 0.27 0.63 1,700 0.28 0.63 1,800
Skarn
TOTAL 2.38 0.77 18,300 4.36 0.91 39,500 6.73 0.86 57,800

Table 6 - MRE Re-stated Assuming 0.5% Sn Cut-off for Skarn and 0.3% Sn Cut-off for Schist

Thus it can be seen that a good size project still exists at these assumed cut-off grades. The
Hämmerlein seam schist and skarn mineralisation, which is either above the water table or just
underneath it, could potentially be a good starter project (18,000t tin), with potential to rapidly
convert the Inferred portion to Indicated by limited de-watering and underground drilling. Even
without this, there is sufficient Indicated category mineralisation (11,000t tin) for a 300,000 tonne
per annum (tpa) operation to start with i.e. 1.72 Mt would provide nearly six years mine life. The
Indicated category mineralisation also has potential to be rapidly converted to Measured category,
as drill density is good and recent work has shown that the Wismut assaying is reliable and, if
anything, conservative (see discussion of underground sampling below).

As well as tin, the deposit also has considerable potential to be able to produce zinc, indium and
magnetite, with other products such as silver, gallium, germanium, lithium, fluorite, abrasives and
road gravel also worth considering. As the majority of these products have not been systematically
analysed for by Wismut, it is difficult to obtain an estimate of their grades and distribution.

However, there are enough zinc assays to obtain a realistic inferred resource estimate. This has
been undertaken by H&S at the request of Saxore but is only classified as inferred due to drill density
and poor QaQc. The total Inferred category estimate is:

Category Mt Zn % Zinc tonnes


Inferred 18.0 1.12 200,400

Table 7 - Tellerhäuser Zinc Mineral Resource Estimate

41
This mineralisation is partially coincident with the tin mineralisation but is also found away from
known tin mineralisation. The average zinc grade within the 0.2% Sn Hämmerlein tin resource is
1.28% Zn and this is considered to be a reasonable number to use for estimating potential by-
products within this seam. More details of this estimate can be found in the separate H&S resource
report.

Indium has only been analysed occasionally by Wismut but there are a significant number of assays
within the database. In order to gain an understanding of the distribution and likely grade for
indium, a default value of 1ppm was entered for all intervals of skarn not assayed for indium (the
schist appears to be devoid of significant indium mineralisation) and this was then block modelled
and an estimate produced by H&S:

Category Mt In ppm Indium kg


Inferred 15.9 127 2,023,000

Table 8 - Tellerhäuser Indium Mineral resource Estimate

Similar to the zinc mineralisation, this mineralisation is not entirely coincident with tin. The average
indium grade within the 0.2% Hämmerlein Sn resource is 96ppm and this is considered to be a
reasonable number to be used for estimating potential by-products within this seam. Interestingly,
the best indium mineralisation is outside the tin resource at Hämmerlein, with 8.2Mt @ 160ppm In
outside the tin resource. The Wismut indium assays have a high detection limit and low precision
(see discussion of repeat channel sampling below) and hence the indium resource should be treated
with caution until additional drill testing can be undertaken. More details of this estimate can be
found in the separate H&S resource report.

Indium is usually produced as a by-product of zinc production and is generally present in the low
10’s of ppm range in zinc deposits, most of which do not extract the indium due to economic
considerations. The indium at Tellerhäuser is much higher grade, appears to be associated more
with copper (although some association with zinc is also observed on a small scale), and is present as
substitution for iron in chalcopyrite (CuFeS2) and as the mineral roquesite (CuInS2) which is
essentially chalcopyrite with all of the Fe substituted by In. Measurements at Tellerhäuser with a
portable XRF machine (pXRF) have shown that chalcopyrite often contains between 1000ppm and
6000ppm In, compared with sphalerite (ZnS) which contains a few tens to hundreds of ppm In.
These measurements have been confirmed by chemical analyses of chalcopyrite rich samples which
have returned up to 6,310ppm In.

As extraction of chalcopyrite, and by analogy roquesite, by flotation is very simple technology (e.g.
most copper mines use this worldwide and it is used to extract copper as an economic by-product at
the Renison Bell tin mine in Tasmania as the first phase of processing), there is every reason to
believe that a high grade copper-indium concentrate can be produced. Discussions with a local
German company that produces indium from photovoltaic scrap suggests that the composition of
such a concentrate will be similar to their starting product and hence that extraction of indium is
feasible from such a concentrate. This company has undertaken preliminary testwork on behalf of
Saxore, with initial results suggesting that at least 48% of the indium goes into solution using off the
shelf acids at room temperature and pressure. Work is ongoing looking at higher temperatures and
pressures and different acids for dissolution.

42
7.3.4 Underground Sampling

In order to examine the accuracy of the previous Wismut assays, a programme of underground
channel sampling was conducted during late 2015. This also had the additional aims of examining
continuity of mineralisation on a mesoscopic (mining) scale and selecting areas for collection of bulk
metallurgical samples (conducted in early 2016). The programme is described in detail in the
separate report “Report on Results of Channel Sampling at Tellerhäuser, June 2016”.

The sampling was designed to duplicate as many of the currently accessible Wismut channels as
possible. A total of 225 samples were collected from 66 channels - all from the Hämmerlein seam
skarn and schist mineralisation from above the water table and away from the active tourist mine
infrastructure. The parts of the seam sampled are shown on Figure 35 below, which shows Wismut
contoured tin grade as a backdrop.

Figure 35 - Channel Sample Locations on Hämmerlein Skarn Grade Map (Yellow 0.1%5 - 0.30% Sn, Green 0.31% - 0.60% Sn,
Orange 0.61% - 0.90% Sn, Red >0.90% Sn)

The channels were not always able to be accurately repeated due to difficulties locating the exact
channels (especially in areas that have been re-excavated after the Wismut sampling) and also due
to not being able to locate the exact base of most of the channels due to fill on the drive floors.
However, given these constraints, enough direct comparisons could be made to show that the tin
assaying by Wismut is repeatable but appears to be underestimating actual tin content by around
12% with a very good correlation between Wismut and Saxore composite assay data (R2 = 0.90).
Hence the Wismut data is considered to be conservative. In fact in more detail, the higher the
mineralisation grade, the more the Wismut assays appear to be underestimating tin grade, by up to

43
50% in some instances, whereas at lower grades (<0.2% Sn) the Wismut data may be overestimating
tin grade.

As anything below 0.2% Sn is well below any economic cut-off, the Wismut data can be considered
to be conservative overall. QaQc data for the Saxore sampling returned excellent results for blanks,
standards and duplicates, showing these tin assays to be accurate and precise.

The zinc and indium assays were not able to be confirmed, as the amount of duplicate assays are not
considered to be statistically valid. The zinc data looks reasonable from the few reliable duplicates
collected and hence the zinc numbers are considered realistic. However, the limited indium data
suggests that at higher grades, Wismut values may be overestimating contained indium whereas at
lower grades, they may be underestimating contained indium. As the background in skarns is
between 10ppm and 40ppm, the fact that 1ppm was used for all unassayed samples (the majority) is
considered to be conservative. Hence any bias due to overestimation due to high grade assays will
most probably be compensated for by the underestimation of skarn background values. However,
the inferred resource estimates for indium should be treated with caution until more sampling of the
higher grade zones can be undertaken.

The schist mineralisation exposed in the main adit returned assays from the Saxore sampling around
45% higher than the corresponding Wismut assays and this data shows an excellent correlation with
and R2 of 0.98. The higher Saxore grade remains to be explained (possibly due to inefficient
dissolution techniques for larger cassiterite crystals during the Wismut work), but again suggests
that the Wismut assays are very conservative. The sampling is in a moderate to high grade schist
block and shows that the mineralisation varies systematically (Figure 36), that various cut-offs could
be used for mining and that grade control should be relatively straightforward.

Fig 36 - Summary Section through Schist Mineralisation, NE Wall of Main Adit (Tin Assays Shown in % Sn - Wismut on Left,
Saxore on Right)

Moderate to high grade skarn mineralisation was sampled in Crosscut 2 and Drive 214 as shown in
Figure 37 below. This data shows good correlation with Wismut sampling with R2 being 0.87 and
0.84 respectively and with Wismut data underestimating tin compared with Saxore data by 4% and
32% respectively.

The data shows that, as interpreted from earlier drive mapping, the main garnet-pyroxene-
amphibole skarn units have relatively low background tin values of between 0.1% and 0.3% Sn and
that higher grade zones occur within the skarns associated with retrograde chlorite-amphibole-
quartz alteration. These rapidly improve in grade to plus 1% Sn (often plus 2% Sn) over very small
distances (Figure 37).

44
Fig 37 - Summary Sections - Crosscut 2, Drive 214 (Tin Assays Shown in % Sn - Wismut on Left, Saxore on Right)

These very high grade zones generally show evidence of coarse cassiterite associated with chloritic
alteration (Figure 38).

Figure 38 - Coarse Cassiterite with associated Quartz-Chlorite Alteration, Cross-cut 2

45
Drives 2-6 and 2-6b are within a low to moderate grade part of the deposit (0.3% to 0.6% Sn) but are
in a part of the deposit that has high visible copper (chalcopyrite) and high indium in pXRF
measurements. Sampling shows that parts of this block actually contain high grade tin
mineralisation and this is most probably associated with the late chloritic overprint (preliminary
petrology has noted chlorite associated with coarse cassiterite here). Figure 39 shows this high
grade tin mineralisation zone cross-cutting the generally low grade skarn. It also shows that these
high grade zones are relatively continuous but have very sharp boundaries.

The indium distribution in this part of the deposit is shown on Figure 40. This clearly shows the
elevated background of indium in the skarn and a sharp drop-off to almost zero in the schist. It also
shows that the high copper assays generally have high indium (as suggested by earlier pXRF testing)
but that relatively high indium can also be found in samples with low copper. Thus more work on
the distribution of indium needs to be undertaken.

Fig 39 - Summary Sections - Drives 2-6 & 2-6b (Tin Assays Shown in % Sn - Wismut on Left, Saxore on Right)

46
Fig 40 - Summary Sections - Drives 2-6 & 2-6b (In/Cu Assays Shown in ppm - Wismut on Left, Saxore on Right)

Drive 2-4 contains a continuous lens of massive sphalerite and gives a good indication of zinc
distribution within a relatively high grade zinc zone (Figure 41). This drive was enlarged subsequent
to the collection of Wismut samples and hence no direct comparison between Saxore and Wismut
assays can be obtained. However, the very high grade and very continuous nature of the zinc
mineralisation can be clearly seen. This drive has very little tin mineralisation and would be sub-
grade. Significant indium and copper mineralisation is present but does not bear any direct
relationship with the zinc mineralisation. As for Drives 2-6 and 2-6b, high copper zones have high
indium but not vice versa.

47
Fig 41 - Summary Sections - Drive 2-4 (Zinc Assays Shown in % Zn - Wismut on Left, Saxore on Right)

Based on the channel sampling, five bulk samples, each consisting of around 10 to 15 tonnes, have
been collected by rock splitting or blasting from:

1. Medium to high grade schist mineralisation from the Main Adit.


2. Medium to high grade skarn mineralisation from Crosscut 2.
3. Low to medium grade skarn mineralisation from Drive 4 (designed to test various pre-
concentration techniques).
4. Low tin grade, high zinc grade skarn mineralisation from Drive 2-4 (designed to obtain high
grade zinc mineralisation).
5. Low to moderate grade tin, zinc, copper and indium mineralisation from Drive 2-6b
(designed to obtain mineralisation with elevated copper and indium).

48
7.4 Metallurgy

7.4.1 Historical Work - General

The reason these deposits were not mined by Wismut is that they were still working on a feasible
processing facility when the re-unification of Germany happened in 1990 and the Soviets walked
away from the joint venture. It is important to look back at their testwork in order to assess the
quality of this work. The following summary broadly reflects the work completed and results
obtained.

Work completed:

1. Initial work at Hämmerlein started in 1971. Almost immediately, the use of gravity was
rejected for the skarn mineralisation based on mineralogical work that suggested about 50%
of the cassiterite is finer than 50 micrometres (μm) and the majority is finer than 100μm.
Gravity techniques at that time were basically restricted to shaking tables and jigs, which are
unable to separate cassiterite finer than around 100μm. Hence, most subsequent work
focussed almost exclusively on flotation.
2. The Hämmerlein schist mineralisation was found to consist of relatively coarse cassiterite
and simple gravity separation resulted in recoveries of 60-70% to a plus 45% Sn concentrate.
No further work was conducted on this type of mineralisation as it was considered that this
was good enough for proof of concept.
3. Laboratory studies continued and a pilot plant was constructed in 1975 capable of 1t/h
throughput. Several thousand tonnes of mineralisation were subsequently tested in this
plant from all deposits and mineralisation types.
4. Research was subsequently conducted by the East German government and various
research organisations between 1981 and 1988.
5. The Hämmerlein mineralisation was concluded to have 80% of cassiterite finer than 100μm
and between 4% and 20% of tin as “isomorphic” tin i.e. non cassiterite or ultra-fine
inclusions.
6. The Tellerhäuser (Dreiberg) mineralisation was concluded to have 40-50% of cassiterite finer
than 100μm and between 16% and 18% of tin as “isomorphic” tin.
7. The final phase of work, completed in 1987, concluded that the following concentrates could
be produced using a minus 70μm grind and gravity/flotation:
a. Hämmerlein (65% skarn, 35% schist): 77% recovery to a 7% Sn concentrate
b. Tellerhäuser (100% skarn): 65% recovery to a 10% Sn concentrate
8. It was suggested that the tin could be upgraded from these 7-10% concentrates to a saleable
product using tin fuming.

All of the previous Wismut mineralogical and mineral processing reports have now been purchased
from Wismut and examined in detail. These are very illuminating and support and expand on our
previous interpretations that this testwork is seriously flawed. In particular:

 Based on over 2000 mineralogical investigations using thin sections and scanning electron
microscopes, it is noted that an average of between 0.15 and 0.20% tin (average 0.186% Sn)
occurs as “soluble tin” which is equated to un-recoverable tin in silicates (amphibole, garnet,

49
pyroxene, malayaite/stokesite (Ca-Sn-silicate) etc). However, this 0.2% is independent of
the actual tin value in the sample. Hence at an average grade of around 0.4% Sn (the
average grade of the metallurgical testwork samples used by Wismut) even if it was possible
to extract 100% of the tin as cassiterite, this would result in a maximum of around 50%
recovery. At a more realistic 75% recovery of cassiterite tin, this would in fact result in, at
best, recovery of (0.75 x (0.4-0.2))/0.4 = 38% recovery. As Wismut report a final recovery of
around 75% tin to a 5-10% tin concentrate, they must have included a substantial amount of
soluble (un-recoverable) tin into the concentrate in order to get such high recoveries. This
would explain the low concentrate grade, as a considerable amount of silicate tin must be
included in the concentrate. Most of the samples that are above 0.5% Sn and especially
those above 1% Sn, tend to show much better recovery, higher tin as cassiterite percentage
and a higher percentage of coarse grained cassiterite. This intuitively should be the case, as
taking off 0.2% Sn from a higher head grade has a much more subdued effect (i.e. recovery
from a 1% Sn head grade should be around (0.75 x (1-0.2))/1 = 60% recovery).
 Analysis of amphibole, garnet and pyroxene show they contain an average of between 0.1
and 0.2% Sn, occasionally higher. This supports the above conclusions and suggests the
concentrates must still contain a significant proportion of these minerals.
 Wismut looked at each panel of mineralisation in Hämmerlein and reported for each block:
total tin grade, soluble tin grade and percentage of tin a cassiterite. This is shown on Figure
42 below and a summary is shown on Figure 43 in graphical format:

Figure 42 – Hämmerlein Mineralisation Panels Showing Total Tin, “Soluble Tin” and % Tin as Cassiterite

50
Figure 43 – Statistics Using Data from Figure 2

Figure 43 is very informative as it shows a direct relationship between cassiterite tin grade
and total tin grade, with cassiterite tin grade equal to total tin grade minus 0.186. Thus it
shows that “soluble tin” is constant and equal to 0.186%, irrespective of actual sample
grade. Wismut have colour coded Figure 2 based on percentage of tin occurring as
cassiterite, however this is meaningless as the ratio is dependent solely on grade so all it is
showing is where the higher grade blocks occur. It would be much more useful to simply
assume that 0.186% of tin occurs in an unrecoverable form and hence subtract this from the
head grade and then assume a “normal” tin recovery of around 75%.
 Wismut looked at grain size of cassiterite and concluded a significant amount is very fine
grained (<40µm). However, this again is dependent on the grade of the sample, with higher
grade samples having significantly higher amounts of coarse cassiterite. This fits with our
own observations that there is a late retrograde mineralising event related to chlorite-
amphibole-quartz alteration that brought in a considerable amount of coarse cassiterite. It
thus appears that the 0.2% “background” tin may include some ultra-fine grained
unrecoverable cassiterite inclusions within silicate minerals and is related to the initial pro-
grade skarnification event. If only higher tin grade blocks are considered, the amount of
coarse cassiterite is almost always high, as shown by the initial mineralogical work
conducted in 1972-73 (Table 9):

51
Sample Lithology Grade % Cass % Cass % Cass
% Sn > 1mm 0.1-1mm <0.1mm
S21 mt skarn 0.44 10 42 48
S22 mt-amph skarn 0.39 0 85 15
S23 mt-amph skarn 0.89 95 4 1
S24 px-amph-ep skarn 0.29 0 50 50
S25 px-amph-mt skarn 0.21 0 50 50
S26 schist 0.23 60 30 10
S27 schist 0.53 25 60 15
Table 9 - Percentage of Cassiterite in Various Size Fractions, Hämmerlein, 1973

This shows that most of the cassiterite is coarser than 0.1mm (100µm) except for very low
grade (below likely cut-off grade) samples, with much of the cassiterite actually being
coarser than 1mm (1000µm) in higher grade samples (>0.5% Sn). Later work conducted in
1979-80 confirmed this, with between 87% and 99% of cassiterite seen to be coarser than
20µm, even though the main conclusion of this work was that only between 13% and 70% of
the tin in skarn occurs as cassiterite coarser than 8µm and that around 50% of the tin occurs
as “soluble tin” or ultrafine grained cassiterite (as we now believe should be expected based
on samples grading 0.2-0.5% Sn!) i.e.:

Mineralisation % of total Grade % cassiterite % cassiterite % cassiterite


Type mineralisation % Sn > 0.5mm 0.02-0.5mm <0.02mm
ga-px skarn 7 0.29 4 94 2
ga-px-amph skarn 4 0.46 5 92 3
amph skarn 5 0.31 2 85 13
amph-mt skarn 31 0.53 4 94 2
mt skarn 34 0.55 6 91 3
schist 19 0.32 23 76 1
Table 10 - Percentage of Cassiterite in Various Size Fractions, Hämmerlein, 1980

This is one reason why Wismut discounted using gravity separation at a very early stage.
However, this conclusion is obviously flawed as it is based on low grade samples which
contain an average of around 0.2% tin in silicates and hence it is impossible to get a
significant amount of the total tin into cassiterite coarser than 0.008 mm. In fact the table
above is actually very encouraging as it shows that between 87% and 98% of the cassiterite
in skarns is coarser than 0.02mm (20µm) and hence once the ~0.2% tin in silicates is
subtracted, good recoveries should be achievable. All of this information supports our own
conclusion that significant quantities of coarse cassiterite are brought into the system in the
last, retrograde, chlorite alteration event, and that good recoveries should be achievable if
0.2% Sn is assumed to be in silicates and is subtracted from the initial grade.
 Some early DMS testwork was undertaken and this confirms that a significant portion of the
“barren” host rock can be removed from the mineralised rock at a very coarse crush size.
For example, at a crush size of 2cm, it is possible to reduce the mass of skarn to
approximately 60% with almost double the grade and only losing between 0.06 and 0.16%
Sn into the light fraction (most of this would probably be “unrecoverable” tin in silicates so
of no concern) using a dense media with a bulk density of around 2.85g/cm3. Interestingly,

52
the crush/grind size does not appear to have any effect on mass reduction or tin recovery.
However, this does not work as effectively for the schist mineralisation.
 Testwork using jigs at a coarse crush (0.5 to 2cm) has shown that mass can be reduced by
33-43% using this technique with losses to tails of between 0.16% and 0.23% Sn (again
probably mainly “un-recoverable” tin). Hence this could be a very cheap alternative to DMS.
 Magnetic separation testwork confirms almost all of the magnetite/martite/haematite can
be separated from the skarns using low magnetic intensity magnetic separation techniques
following a minus 0.5mm crush/grind. Most of the magnetite reports to the magnetic
fraction, which averages plus 60% Fe (up to 68% Fe) and hence is a potentially saleable iron
ore. This magnetic concentrate contains between 0.08% and 0.3% Sn and averages just
under 0.2% Sn. From our recent work we know that most of this tin occurs in the form of
ultra-fine grained cassiterite within magnetite and would most probably be largely un-
recoverable anyway. The tin content of the iron concentrate may be a concern (this is being
confirmed by talking with smelters) but no other deleterious elements are known to occur.
 Some early gravity testwork was undertaken and this has shown that much of the cassiterite
tin can be concentrated using conventional techniques such as spirals and shaking tables.
 A final pilot scale plant was constructed consisting of crushing/grinding the entire sample to
100% passing 0.5mm (creating obvious overgrind of cassiterite) followed by spirals and
shaking tables, followed by ball milling all tails to 100% passing 70 µm (again creating
overgrind), followed by sulphide flotation followed by cassiterite flotation. Interestingly, no
DMS, jigs or magnetic separation were utilised - this would have had the effect of creating
significant problems in the gravity circuit due to “clogging” by magnetite which has an s.g.
not much lower than cassiterite. Figure 44 below shows the final flow sheet used for the
testwork:

Figure44: Pilot Plant Flow Sheet, Hämmerlein

53
This was used for multiple tests using slightly different configurations. Based on these
preliminary tests, three “production” samples were put through the final configuration:
skarn, schist and mixed skarn schist. Results of this show that significant amounts of tin are
recovered to a moderate to high grade concentrate using the shaking tables only. For
example using skarn only, 23.4% of the tin reports to the high grade shaking table
concentrate grading 8.8% Sn with 3.8% of total tin reporting to a concentrate grading 17.4%
Sn. In this same sample, only 32.9% of the tin reports to a flotation concentrate grading
2.88% Sn. Schist mineralisation is even better with 41.9% of the tin reporting to a gravity
concentrate grading 21.9% Sn including 17.5% reporting to a 41.1% Sn concentrate. In this
sample, 28.8% of the tin reports to the flotation concentrate which only grades 5.65% Sn.
Interestingly, the mixed ore provides the best results with 17.6% of the tin reporting to a
gravity concentrate grading 50.9% Sn with flotation recovering 28.7% of the tin to a 2.85%
Sn concentrate. These results need to be considered in light of their low grade and the
problems with this discussed above. However, based on these results, modern gravity
techniques would seem to provide an excellent opportunity for successful processing of the
mineralisation.

In summary then, the previous testwork can be seen to be seriously flawed from the start for the
following reasons:

1. It was shown that approximately 0.2% tin (average 0.186% Sn) always occurs as un-
recoverable tin, mainly as tin silicates and possibly some ultrafine grained cassiterite which is
variably referred to as soluble tin or isomorphic tin.
2. This was not however, carried through to later work and hence recovery percentages based
on low grade samples (average around 0.4% Sn) could never hope to obtain better than
about 30-50% recovery unless some of the “un-recoverable” tin is included in the
concentrate. Hence using percentage figures is totally misleading and is a common and
consistent problem with all this work.
3. Gravity, DMS and magnetic separation techniques were rejected, as the tails were running
around 0.15% to 0.20% Sn, thus making recoveries unacceptable. However, based on the
above two points, it is clear that this tin is probably unrecoverable anyway.
4. A gravity circuit (one spiral followed by three shaking tables) was built into the pilot plant
despite this. However, magnetite was not removed prior to gravity separation and this
would have made the gravity separation very inefficient (density of cassiterite is about
7.0g/cm3 and density of magnetite is about 5.2g/cm3). Even given this, significant tin
recoveries to moderate to high grade concentrates were achieved.
5. All material was crushed and ground in a rod mill to 100% passing 0.5mm prior to any
treatment. This would have created significant overgrind and the fines should have been
removed and treated separately at each stage of comminution.
6. All rejects from gravity were reground to minus 70 µm and sent to sulphide flotation
followed by cassiterite flotation.
7. Flotation of sulphides did not remove any appreciable tin mineralisation and hence clean-up
should be simple.

54
8. Cassiterite flotation recovered most of the remaining tin but to a very low grade concentrate
(2.5% to 3% Sn) due to wanting to keep the tailings low. If this were changed to multi-G
gravity separation (Falcons etc.), this should substantially improve both recovery and
concentrate grade, as tin in silicates would not be “recovered”.
9. If it is assumed that around 0.2% of tin is un-recoverable, then any tailings grading 0.2% tin
are not effectively losing any “recoverable tin”. Hence, even the magnetic separation stage
should not lose any significant “recoverable tin”.
10. If DMS/jigs are used to remove around 40% of the total mass, with losses to the light
fraction of around 0.2% Sn, and then magnetic separation is used to remove a further 25% of
the original mass grading around 0.2% Sn, then around 65% of total mass will have been
removed, grading around 0.2% Sn, most of which is probably “un-recoverable”. This will
substantially reduce the mass needed to be put through the main milling complex to around
35% of the mass mined and at around double the grade mined, thus greatly reducing
processing costs and making gravity separation much more efficient (as the magnetite will
have been largely removed and the next heaviest minerals are much less dense, having
densities of less than 4.0g/cm3, with the majority less than 3.5g/cm3).

Hence it is concluded that it should be possible to devise a workable processing route based on
these results and ongoing testwork will focus on these areas.

7.4.2 Dense Media Separation

As noted in the discussion above, excellent potential exists to pre-concentrate the mineralisation
prior to milling, thus reducing throughput through the main milling circuit and significantly
increasing the grade put through that circuit. This would have the added benefit of reducing the
amount of tailings significantly and thus making surface permitting quicker and simpler.

Various options exist for pre-concentration including dense media separation (DMS), optical sorting,
XRF sorting, jigs etc. Some test work has been conducted on DMS and the results of this work are
very encouraging.

A report on work conducted in 1971 returned the following results:

Table 11: Dense Media Testwork Summary, from 1971

This suggests that the schist mineralisation can be upgraded to a 9.2% Sn concentrate with 84% tin
recovery from a 0.19% Sn head grade by simply crushing to 2mm and using a dense media with a
specific gravity of 3.3g/cm3.

55
Similarly, it suggests that the low magnetite-sphalerite skarn can be upgraded to a 27% Sn
concentrate with 81% tin recovery from a 1.4% Sn head grade by crushing to 20mm and using a
dense media of 3.3g/cm3.

The magnetite/sphalerite skarn cannot be upgraded to such an extent because both magnetite (4.9-
5.2g/cm3) and sphalerite (3.9-4.2g/cm3) have specific gravities considerably greater than 3.3g/cm3
(compared with the other skarn minerals such as garnet (3.1-4.4g/cm3), amphibole (2.9-3.2g/cm3)
pyroxene (3.2-3.6g/cm3) and marble plus other calc-silicates (2.4-2.9g/cm3)) and both make up a
significant proportion of the sample (~50%). However, the testwork did show that the head grade
can be approximately doubled with a tin recovery of 86% using either a 2mm or a 20mm crush and a
dense media of 2.9g/cm3 or 3.3g/cm3. Cassiterite itself has a specific gravity of about 7.0-7.1 and
hence should report to the heavy fraction if sufficiently liberated.

If a magnetite-sphalerite-cassiterite concentrate can be produced, it should then be a relatively


simple process to separate the sphalerite (and any other sulphides) out using flotation and the
magnetite using magnetic separation.

A preliminary test on several random samples from the open part of the Hämmerlein Seam has been
undertaken by Saxore in order to verify these early results. While the samples collected were sub-
optimal (due to most of them being sub-grade) they did confirm that in some cases (high grade
schist in this case) a very high grade concentrate can be produced by a coarse crush and simple
dense media separation and that in all cases apart from the magnetite rich skarn, grade can be
approximately doubled and mass approximately halved.
Lithology Crush Calc Grade Fines (<0.09mm) <2.9 sg 2.9-3.3 sg >3.3 sg >2.9 sg
(% Sn) Mass % Tin % Grade (% Sn) Mass % Tin % Grade (% Sn) Mass % Tin % Grade (% Sn) Mass % Tin % Grade (% Sn) Mass % Tin % Grade (% Sn)
Schist <2mm 1.57 12.80 10.90 1.34 72.8 10.2 0.22 12.0 17.8 2.34 2.4 61.1 39.30 14.4 78.9 8.60
<20mm 1.34 3.40 3.02 1.15 84.8 32.2 0.51 13.6 37.1 3.66 1.5 27.6 24.00 15.2 64.7 5.73
Garnet Skarn <2mm 0.16 13.60 14.08 0.14 19.5 1.2 0.01 23.7 12.1 0.08 56.8 72.6 0.20 80.5 84.7 0.16
<20mm 0.15 1.60 2.01 0.18 10.2 4.2 0.06 33.2 16.0 0.07 56.6 77.8 0.20 89.8 93.8 0.15
Sulphide Skarn <2mm 0.36 15.00 16.40 0.33 37.9 5.3 0.05 20.6 8.1 0.14 41.5 70.2 0.60 62.1 78.3 0.45
<20mm 0.32 2.40 3.41 0.45 32.1 5.9 0.06 26.3 13.8 0.17 41.6 76.9 0.60 67.9 90.7 0.43
Amphibole Skarn <2mm 0.27 20.60 20.30 0.21 52.7 15.7 0.08 22.4 17.5 0.21 24.9 46.5 0.50 47.3 64.0 0.36
<20mm 0.21 2.00 2.71 0.28 51.4 17.1 0.07 29.7 35.3 0.25 18.9 44.9 0.50 48.6 80.2 0.35
Magnetite Skarn <2mm 1.37 16.50 17.69 1.23 8.8 3.0 0.47 6.0 4.9 1.12 85.2 74.4 1.20 91.2 79.3 1.19
<20mm 1.16 2.20 2.14 1.10 4.7 1.6 0.4 8.1 5.8 0.84 87.2 90.4 1.20 95.3 96.2 1.17
Pyroxene Skarn <2mm 0.49 18.30 20.98 0.46 31.9 6.5 0.1 26.1 21.2 0.4 42.0 51.3 0.60 68.1 72.5 0.52
<20mm 0.44 1.90 2.46 0.56 28.3 1.3 0.02 30.2 30.1 0.44 41.6 66.1 0.70 71.7 96.3 0.59

Table 12: Dense Media testwork Summary, Recent Saxore Results

It can be seen from the above table that the high grade schist ore can be concentrated from 1.5% Sn
to 8.6% Sn in 15% of the original mass at a recovery of 79% using a 2mm crush and a 2.9g/cm3 dense
medium. Alternatively, it can be upgraded to 39.3% Sn in 2.4% of the original mass at a recovery of
61% using a 2mm crush and a 3.3g/cm3 dense medium.

It can also be seen that the majority of the skarn samples, even though sub-grade and hence
expected to have poor recovery, can be significantly upgraded by simply crushing to 20mm and
using a 3.3g/cm3 dense medium. This returns the following results:

 Garnet skarn upgraded from 0.15% Sn to 0.20% Sn in 57% of the mass with 78% recovery
 Sulphide skarn upgraded from 0.32% Sn to 0.60% Sn in 42% of the mass with 77% recovery
 Amphibole skarn upgraded from 0.21% Sn to 0.50% Sn in 19% of the mass with 45% recovery
 Pyroxene skarn upgraded from 0.44% Sn to 0.70% Sn in 42% of the mass with 66% recovery

56
Note that recovery for the amphibole skarn can be improved by using a 2.9g/cm3 dense medium
which returns an 80% recovery. As better grade samples are collected and tested and the technique
is finessed, it is probable that significant improvements on this preliminary work can be expected.

In all cases except the magnetite skarn, the amount of tin in the lighter than 2.9g/cm 3 fraction is
negligible. Given the previous discussion on tin in silicates, the low tin in tails is the main factor of
interest from this work. Additional work is currently underway using more representative and
better grade samples.

7.4.3 Recent Advances

As gravity techniques were rejected based on flawed assumptions in 1975 and hence only limited
further testwork on these techniques undertaken, recent advances in this form of mineral
processing have not been considered. These include:

 Knelson Concentrators – developed for the iron ore industry but now in use in tin mines in
Australia, can separate cassiterite down to about 30 microns.
 Kelsey Jigs – previously used at the Renison Bell tin mine in Tasmania, Australia since the
early 1990’s. Use a centrifugal force of up to 50G. Can be used to separate cassiterite as
fine as 5-10 microns.
 Falcon Centrifugal Concentrators – can develop a centrifugal force of up to 600G and thus
can be used to separate very fine cassiterite. Various versions exist and the best can
separate cassiterite as fine as 3 microns. These are used extensively at the Renison Bell tin
processing plant in Tasmania.

Based on these advancements in gravity separation techniques and recent detailed testwork on
similar styles of skarn hosted tin mineralisation elsewhere (e.g. Renison Bell, Consolidated Tin’s Mt
Garnet deposits), the previous metallurgical testwork should be considered virtually obsolete.
Revised testwork focussing on gravity separation techniques (including more DMS testwork), but
also including magnetic separation and flotation techniques (designed to obtain separate iron, zinc,
copper/indium and silver concentrates and to improve overall tin recovery) is proposed. This work is
currently in progress.

7.4.4 Pyro-metallurgy

A worst case scenario, assuming the final concentrates obtained by SDAG Wismut (~65-77% recovery
to a 7-10% Sn concentrate) are the best that can be produced, has been examined to see if such a
low grade concentrate can be economically processed using pyro-metallurgical techniques.

An independent desk top scoping study undertaken by Nic Barcza and Mike & Greg Freeman, in
conjunction with Mintek South Africa, suggests that such a concentrate can be economically treated
in a commercially available pyro-metallurgical DC furnace to produce a tin matt grading plus 95% Sn
at the then tin price of US$23,500/t (95% payable, US$200/t processing charge) and making some

57
educated guesses regarding mining and processing costs to produce the concentrate and the capital
costs required. This report is now out of date but is available separately if required.

One of the significant advantages of the Tellerhäuser deposit is that it is located within a short
distance of several existing smelting facilities with active emission permits. One is located at Aue,
approximately 15km (20km by road) from the deposit, another is located in Freiberg, approximately
60km (85km by road) from the deposit.

7.4.5 Other Potential Products

Potential to economically extract the other products and thus add value is considered to be good:

 Magnetite should be easily separated using wet and dry magnetics and should be readily
saleable as either iron ore to steel mills or, if the tin content makes this not feasible, as a
coal washing product. Several low quality coal mines are located within a 100km radius of
Tellerhäuser both in Germany and the Czech Republic. The closest is only 30km away in the
Czech Republic.
 Zinc occurs as coarse, massive sphalerite and is directly associated with the tin skarns. This
should be able to be simply separated to a mixed sulphide flotation product.
 Silver appears to be associated with zinc and other sulphides (notably arsenopyrite) and
should report to a sulphide concentrate, although recoveries are unknown.
 Indium appears to be mainly associated with copper sulphides and to a lesser extent with
zinc sulphides and other minerals yet to be confirmed. As such, a separate copper sulphide
concentrate should be able to be floated from the mixed sulphide concentrate or else
floated separately before sphalerite is floated. This should be a very high grade indium
concentrate (copper of negligible relative value) and should be able to be treated as a
primary indium concentrate. As noted in section 7.3, testwork is currently being conducted
on a copper-indium concentrate to ascertain how efficiently indium can be recovered.
 Gallium and Germanium may be recoverable from the other concentrates but as the
distribution of these metals is not known, testwork is planned to ascertain their precise
mode of occurrence.
 Abrasives are present as garnet (mainly andradite so not useful for high abrasive waterjet
cutting), pyroxenes and amphiboles (mainly hornblende) and a mixed concentrate of these
may be saleable as a low grade abrasive. These should be easily separated after a coarse
crush and enquiries are currently being made as to their potential value.
 Road gravel is a very real option for reducing volume of waste dumps on site, even if this
product only pays for its own removal. Studies to date suggest that DMS/jigs could readily
produce a product with suitable characteristics (low sulphide etc).

While these by- and co-products could make processing more complex, they have the potential to
add considerable value, especially indium, and hence ongoing work on their extraction is considered
to be a high priority. In addition there is a large benefit to keeping tailings and waste rock volumes
as low as possible so even if products such as road gravel and magnetite can pay for their own
removal this would be of significant benefit to the operation as a whole.

58
7.5 Environmental & Socio-Economic Considerations

The Erzgebirge is a historical mining area with mining having proceeded almost continually between
1168 and 1990. Some small-moderate scale mining activity is currently being re-established
(principally fluorite and limestone at this stage) and there is a strong local desire to re-establish the
district as a mining centre. This is reflected at State and Federal level with the state minister and
secretary responsible for mining making positive comments that the state government wants to re-
establish the district as a major mining centre (various newspaper articles).

At Federal and EEC level, directives for Germany and the EU to become more self-sufficient with
regard to natural resources is also highly encouraging. Tin has recently been added to the critical
mineral list in Germany and joins indium, gallium and germanium, meaning there is a strategic
reason for the government to help progress this project.

However, as the project is located in Europe and is in a popular tourist area (hiking, cross country
skiing etc), some opposition from green oriented parties can be envisioned. It is considered likely
that this will not stop the project but that best environmental practices will be required and the
project will have a high level of scrutiny.

The local environment consists of regrowth pine forest (with some deciduous patches) which is
periodically harvested by the forestry department. Numerous small villages exist in the region and
the main form of income is tourism. The area is in a relatively low income district with relatively
high unemployment and, although it is popular with tourists, these are mainly from within Germany
and in particular what used to be East Germany and average per capita spend is low.

A mining lease for uranium covers most of the known mineralisation. This is owned by Wismut
GmbH, a government controlled organisation now charged solely with the clean-up and
rehabilitation of the old uranium mines. At Tellerhäuser, Wismut runs a small plant that de-
mineralises water draining from the old underground workings. This water originally had elevated
levels of radon, radium and heavy metals (arsenic, bismuth, cadmium etc) but now is only slightly
enriched in heavy metals, principally arsenic. It was previously operated as an active chemical plant
but was then re-designed as a passive biological (wetland) plant. It was recently re-configured yet
again back to an active chemical plant as the biological plant was not operating to specifications.
Water flows are relatively low and mineral levels in the water are also low (only slightly above
acceptable background according to EU guidelines). However, it is estimated that about one million
cubic metres of water is in the current mine openings below the adit level and hence any de-
watering will be a substantial undertaking. It would be necessary to significantly increase the
current capacity (~60t/h) in order to achieve dewatering within a reasonable time frame. Assuming
a 1 year timeframe and 1M m3 of water, a capacity of around 125t/h would be required.

Discussions with the MD of Wismut suggest that the ML should not interfere with ongoing
exploration and in fact they are willing to help with underground sampling as required. Eventually, if
large scale mining commences, it may be necessary to take over running of the water treatment
facilities. However, some water treatment facilities will be required in any case, as any water
released into the environment must be below EU prescribed levels of contaminants. The existing
plant could be upgraded and ongoing running partly funded by the government (Wismut), hence this
is not seen as a major obstacle.

59
Potential exists to backfill the old uranium workings with waste rock and tailings, thus providing a
unique opportunity for below ground disposal (which may be an advantage due to environmental
considerations) and closing off the radioactive contamination source permanently.

It may also be possible to use some of the tailings to produce a paste fill for ground support in the
active workings. However, ultimately disposal of tailings and waste rock is likely to be an issue as
above ground disposal will be subject to very close environmental scrutiny.

Thus, while environmental considerations are significant, they are considered to be manageable,
with disposal of waste rock and tailings and release of mine water into the environment being the
two most critical elements.

Wismut has verbally suggested they are willing to sell the existing mine infrastructure for a nominal
amount (€1) but that would mean taking on the existing environmental liability. This will need
negotiation as a bond will be required. However, it is likely that support will be obtained from all
levels of government (local, state and federal) as all want mining to be re-established, jobs to be
created and supply of critical metals to be assured. As such, potential exists to obtain government
guarantees and indeed co-funding for ongoing environmental rehabilitation (especially water
treatment) as part of the ongoing liability is due to previous mining activity. There is in fact a
precedent set with a recently opened fluorite mine having its environmental bond partially
underwritten by the state government.

7.6 Exploration Potential

Potential to significantly increase the life of any potential operation at Tellerhäuser can be seen
within the Breitenbrunn licence.

Potential to increase tonnes and grade can be seen in the following areas:

1. Peripheral to existing known mineralisation: Wismut has identified a surface area greater
than that of, and peripheral to, the known mineralisation with potential for extensions of
this mineralisation (Figure 45). They had a rule of thumb that favourable lithologies (marble,
calc-silicate etc) within 800m of the granite contact were likely to be mineralised. In
addition, several parts of the mineralisation appear to terminate very abruptly along linear
zones. It is highly likely that these are faults that have offset the mineralisation and
potential for discovering the offset portions must be considered high priority targets (e.g.
Figure 45).
2. Additional skarns below existing known mineralisation: The area at depth below the known
mineralisation has not been adequately tested for addition skarn layers. If present these
could have a higher grade, as it appears that grades may improve in proximity to the granite
source.
3. Between existing known deposits: The area between the known skarn deposits appears to
have been poorly tested but the geological unit hosting the skarns appears to be relatively
continuous. Assuming the source granite is not too deep, potential for additional

60
mineralisation in this unit must be considered to be high. Some drilling has intersected tin
mineralisation throughout this unit as shown on Figure 46.
4. At Breitenbrunn and Antonsthal: These two deposits have only been drill tested from
surface and at a very wide drill spacing (Figures 46). Breitenbrunn has been drilled on an
approximately 200m x 200m pattern and an historical resource of 22.3Mt @ 0.25% Sn
(55,000t tin) has been estimated in GDR times. This is classed as Delta category and it is
doubtful it would be able to be converted to any JORC compliant category without some
additional drilling or underground sampling. Antonsthal has been drilled on a roughly 400m
x 800m pattern and historical resources of 14.7Mt @ 0.19% Sn (28,000t tin) and 6.2Mt @
0.37% WO3 (23,000t WO3) have been estimated in GDR times. These estimates are unlikely
to be converted to JORC compliant resources without considerable additional drilling. Hence
the historical resource estimates at both these prospects are best considered as geological
targets rather than actual resource estimates. It can be seen from an analysis of the detailed
drilling at Tellerhäuser that drilling at 200m-800m spacing would be unlikely to obtain a
realistic idea as to the grade and size of mineralisation, and in fact is likely to miss the higher
grade portions of the Hämmerlein, Dreiberg and Zweibach seams, which are generally 200m
x 200m in lateral extent. Hence the low grades noted in these deposits may not be realistic
and they are considered to represent good targets for potential higher grade mineralisation.
In fact recent information on underground sampling and drilling at Breitenbrunn shows good
potential for several 200m x 200m pods of plus 0.5% tin mineralisation (Figure 47).
5. Tungsten: Tungsten mineralisation intersected in the very wide spaced drilling at Antonsthal
suggests reasonable tonnages of relatively good grade mineralisation (averaging 0.37% WO3
and up to as high as 1.64% WO3 over widths of between 0.4m and 3.5m). No direct search
for tungsten has been made and this could represent a significant target in its own right.
6. Greisen style mineralisation: Potential for simple greisen style mineralisation, especially
related to cupolas associated with the mineralising granite is a real target that remains
untested within the Breitenbrunn EL.

Hence, considerable upside can be seen and the large metal inventory identified to date should be
able to be grown, and grade improved, by additional exploration. In particular the skarn bearing
horizon close to the mineralising granite contact is considered a prime exploration target.

61
Abrupt termination of
mineralisation - interpreted
fault. Potential for discovering
fault offset continuations.

Figure 45 – Potential Mineralisation Peripheral to Known Mineralisation As Identified by Wismut

62
Figure 46 – Surface Drilling – Breitenbrunn Licence

63
Figure 47 - Breitenbrunn Prospect Summary Plan

64
8.0 Research

As the Tellerhäuser deposit is large, unique and readily accessible, it is of interest for study by
various research organisations.

There are currently 2 major projects active, one being funded by the EU (FAME Project) and the
other by Freiberg University (AFK project). In addition, a proposal for a third project is currently
being prepared focussed on pyrometallurgy and hydrometallurgy.

FAME (Flexible and Mobile Economic Processing Technologies) - 18 Partners throughout Europe
(including the Natural History Museum of London), co-ordinated by Wardell Armstrong
International in the UK

This project is examining technology for processing mineralisation using transportable methods that
may be used for small deposits, environmentally sensitive areas or for a series of deposits that may
not be within easy trucking distance of each other. It is also looking at technology for treating
metallurgically complex deposits that have been considered non-processable in the past. The
ultimate aim of the project is to enable economic processing of currently uneconomic deposits
within the EU.

Saxore has signed an NDA with the project and attended a progress meeting in Nancy, France. At
this meeting it was agreed internally to Saxore becoming a full member of the project with initial
funding of around €30,000 to be provided for Saxore staffing and support costs. Additional funding
should be forthcoming as the project progresses. Tellerhäuser is one of the main deposits being
examined and thus will have a significant portion of the €7.5M budget allocated to it. Thus apart
from Saxore benefiting from direct funding of staff and support from this project, it is also obtaining
several million Euros of mineralogical, metallurgical and mineral processing testwork at no cost.

Work has just started on this project and initial Qemscan results have been provided to Saxore from
the initial characterisation work.

Twenty four samples of the schist mineralisation were examined (average grade 1.75% cassiterite)
and this shows that most of the tin occurs as relatively coarse grained cassiterite as expected with
grain size varying from 10 µm up to greater than 4mm.

Eighteen samples of skarn mineralisation were examined (average modal content of 0.05%
cassiterite), all from the low grade (sub cut-off grade) sphalerite chamber. This shows that a
considerable portion of the tin in this chamber occurs as non cassiterite tin, as expected. As the
average grade of the samples was actually 0.25% Sn, this again shows that around 0.2% Sn occurs as
tin in silicates. Malayaite was estimated at 0.02% volume percent. Work on higher grade samples is
currently in progress.

One ton of sample from each of the five bulk sample locations has been provided and detailed work
is expected to commence shortly on these bulk samples.

AFK (Processing of Fine Grained Complex Ores) - 6 Partners (mainly local), co-ordinated by the
Helmholtz Institute in Freiberg

65
This project is being co-ordinated by the Freiberg based Helmholtz Institute for Resource Technology
and includes researchers from Freiberg University (TU Bergakademie Freiberg), Aachen University
(RWTH Aachen) and UVR-FIA (Freiberg based metallurgical testing facility). Its aim is to develop
technologies for processing of complex and/or fine grained mineralisation.

Bulk samples have been delivered to the local laboratory (UVR/FIA) for comminution. This will
consist of crushing the entire sample to around 6cm, followed by homogenising and sub-sampling.
As a result, Saxore will not have to pay for this aspect of the work or the various rock strength
measurements (Bond Work Index etc).

Ongoing work has resulted in information regarding the deportment of tin as shown in the figure
below:

Figure 53 - Deportment of Tin in Various Rock Types by Grade and Mineralogy

This shows that around 0.1-0.2% tin occurs as tin in silicates, as expected. It also shows that the
highest tin and tin as cassiterite percentage occurs within chlorite altered areas, again as expected
from our own underground mapping and sampling.

A recent update shows modal mineral contents from Drives 2-4, 2-6b and 4. This shows a strong
positive correlation between chlorite (green) and cassiterite (red) and a generally negative
correlation between garnet-pyroxene (yellow) and cassiterite (Fig 54) as previously interpreted, thus
supporting the idea that the late chlorite overprint has introduced much of the tin and especially
most of the tin as coarse cassiterite. These are still from relatively low grade drives (2-4 average

66
0.12% Sn, 2-6b average 0.58% Sn and 4 average 0.39% Sn) and it can be seen that the best
correlation (and highest average chlorite content) is in the highest grade chamber - 2-6b. Work on
the higher tin grade chambers is in progress.
Mineral 2-4_1 2-4_2 2-4_4 2-4_5a 2-4_5b 2-4_6 2-4_7a 2-4_8 2-4_9I 2-4_9II 2-4_10 2-4_12 2-4_14 2-6b_1 2-6b_2 2-6b_5I 2-6b_6 2-6b_7 2-6b_8 2-6b_9 4_1 4_2bI+II 4_3
Quartz 32.6 36.5 3.1 7.4 4.3 33.6 1.3 8.5 6.4 3.2 6.0 5.3 11.3 16.6 8.1 16.7 7.5 23.5 23.0 11.2 35.6 27.8 20.4
Feldspar 39.0 0.0 0.0 0.9 0.1 41.3 0.0 0.4 0.0 0.0 0.3 3.6 0.0 20.7 43.6 1.0 9.3 0.4 23.7 30.5 55.4 28.4 51.0
Other Garnet 0.2 0.0 2.2 0.1 0.0 0.0 54.7 38.8 3.6 26.5 0.6 8.5 0.4 0.0 0.8 2.3 0.5 0.1 0.5 0.2 0.0 0.1 0.1
Almandine 0.1 0.6 2.8 0.0 0.0 0.3 0.0 0.1 0.1 1.3 0.2 2.5 0.1 4.0 0.6 0.3 0.8 0.4 0.6 0.8 0.1 1.3 0.4
Epidote 0.2 0.0 3.7 0.4 0.0 0.4 0.0 0.0 0.0 0.4 0.1 16.7 0.0 0.4 10.2 0.6 0.0 0.0 0.3 3.4 0.1 0.3 0.3
Sn-Epidote 0.2 0.0 5.9 0.2 0.0 0.1 0.0 0.0 0.0 1.9 0.2 4.2 0.0 0.0 2.0 0.5 0.0 0.0 0.1 0.6 0.0 0.2 0.1
Sn-Amphibole 4.4 0.5 58.1 0.3 0.0 0.3 0.2 7.5 31.4 52.6 23.5 24.4 16.6 0.1 6.3 2.2 0.5 0.8 2.2 14.7 0.0 7.0 0.6
Amphibole 0.5 0.0 0.4 0.1 0.0 0.1 0.0 0.4 3.0 0.7 4.0 2.4 36.6 0.0 2.2 0.0 0.0 0.0 0.0 0.7 0.0 0.0 0.0
Pyroxene 0.0 0.0 0.0 0.0 0.0 0.3 0.1 14.1 44.1 2.1 0.2 23.4 24.6 0.0 5.7 0.0 0.0 0.0 0.0 0.6 0.0 0.0 0.0
Chlorite 8.0 33.9 8.7 0.6 0.1 0.9 0.2 0.9 1.3 4.3 2.7 2.5 0.2 2.9 3.0 4.9 24.2 35.6 40.6 12.0 1.8 15.5 5.0
Biotite 1.9 0.0 5.9 5.7 0.1 3.8 0.0 0.1 0.0 0.0 3.5 0.1 0.1 16.6 2.0 9.0 0.2 0.6 3.2 1.8 0.1 8.1 6.0
Muscovite 3.8 0.0 0.0 0.0 0.0 17.7 0.0 0.1 0.0 0.0 0.0 0.0 0.0 36.5 3.0 0.7 0.0 1.8 0.2 1.1 5.1 1.8 13.4
Titanite 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.6 0.0 0.0 1.2 0.0 0.0 0.0 0.1 0.8 0.0 0.0 0.4
Sn-Titanite 0.1 0.0 0.1 0.1 0.0 0.1 0.0 0.2 0.6 0.1 0.1 0.6 0.0 0.0 0.4 0.0 0.0 0.0 0.0 0.6 0.1 0.2 0.5
Stokesite 0.0 0.0 0.7 0.1 0.0 0.0 0.0 0.0 0.0 0.2 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.0
Cassiterite 0.1 1.2 0.0 0.0 0.2 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.1 0.0 0.0 0.8 0.7 3.8 0.3 0.1 0.0 0.7 0.0
Rutile 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
Sn-Rutile 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.1 0.8 0.0 0.0 0.0 0.4
Iron Oxide 5.4 8.1 2.9 75.1 81.5 0.2 19.5 24.5 5.6 2.2 52.1 2.6 7.2 0.2 7.0 39.9 0.1 3.8 0.1 14.1 0.1 3.6 0.1
Sn-Iron Oxide 0.9 0.7 1.3 8.5 13.2 0.0 1.3 2.1 0.7 0.5 5.6 0.3 0.6 0.0 0.8 5.8 0.0 0.5 0.0 1.3 0.0 1.3 0.0
Scheelite 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
Native Arsenic 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
Fluorite 0.1 10.2 0.0 0.2 0.2 0.0 0.0 0.0 0.0 1.6 0.0 0.2 0.0 0.0 0.0 4.0 29.6 24.6 0.7 2.7 0.6 2.3 0.6
Calcite 1.3 0.3 2.2 0.1 0.0 0.0 0.0 2.0 2.8 0.8 0.1 1.1 1.9 0.0 2.2 10.8 24.7 1.2 3.3 0.9 0.1 0.2 0.3
Roquesite 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
In-Sphalerite 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
Sphalerite 0.0 3.8 0.0 0.0 0.1 0.0 21.8 0.1 0.1 0.0 0.0 0.8 0.0 0.0 0.0 0.0 0.2 0.0 0.0 0.0 0.0 0.0 0.0
Chalcopyrite 0.0 0.3 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.1 0.0
Pyrite 0.0 3.4 0.0 0.0 0.0 0.0 0.6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.5 0.0 0.0 0.6 0.0 0.0 0.0
Bornite 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
Arsenopyrite 0.0 0.0 1.2 0.0 0.0 0.0 0.0 0.0 0.0 0.8 0.0 0.0 0.0 0.0 0.0 0.2 1.0 2.8 0.0 0.8 0.0 0.1 0.0
Other 0.9 0.1 0.8 0.1 0.1 0.7 0.2 0.2 0.2 0.8 0.5 0.3 0.2 1.8 0.8 0.1 0.0 0.0 0.3 0.4 0.7 0.7 0.4

Figure 54 - Modal Mineral Contents of Selected Samples (showing good correlation between chlorite and cassiterite)

67

Potrebbero piacerti anche