Outline for PGE Report

MINNESOTA GEOLOGICAL SURVEY Open-File Report: OF-06-3

FINAL REPORT ON THE GEOLOGY, GEOCHEMISTRY, GEOPHYSICAL ATTRIBUTES, AND PLATINUM GROUP ELEMENT

POTENTIAL OF MAFIC TO ULTRAMAFIC INTRUSIONS IN MINNESOTA, EXCLUDING THE DULUTH COMPLEX:

PROJECT SUMMARY

Mark A. Jirsa(1), James D. Miller, Jr.(1), Mark J. Severson(2), and Val W. Chandler(1)

With contributions by Terrence J. Boerboom(1), Richard S. Lively(1), Merida J. Keatts(3), and Daniel Holm(3)

1) Minnesota Geological Survey, University of Minnesota, Twin Cities 2) Natural Resources Research Institute, University of Minnesota, Duluth 3) Kent State University, Ohio

NOTE: This open-file report has not been reviewed to conform strictly to editorial standards of the Minnesota Geological Survey

Jirsa and others, 2006 MGS OF-06-3

Table of Contents

EXECUTIVE SUMMARY ............................................................................................................... 3 INTRODUCTION............................................................................................................................. 4 Rationale .................................................................................................................................... 4 Objectives and Methods ............................................................................................................. 5 GEOLOGIC SETTING..................................................................................................................... 5 PROJECT RESULTS....................................................................................................................... 11 Inventory of Intrusions .............................................................................................................. 11 Geochemical Database .............................................................................................................. 11 Worksheet GDT ................................................................................................................. 11 Worksheet CIPW ............................................................................................................... 13 Geochronologic Study............................................................................................................... 14 Geophysical Data ..................................................................................................................... 15 Attributes of Select Intrusions ................................................................................................... 16 Intrusion Name................................................................................................................... 17 County................................................................................................................................ 17 DDH or OC ........................................................................................................................ 17 Province ............................................................................................................................. 17 Intrusion Type .................................................................................................................... 17 Age .................................................................................................................................... 19 Geologic Setting................................................................................................................. 20 General Lithologic Description.......................................................................................... 21 Normative Rock Types ...................................................................................................... 21 Internal Differentiation....................................................................................................... 23 Metamorphism/Alteration .................................................................................................. 23 Geochemical Attributes...................................................................................................... 24 Alk/Thol/CA/Kom ...................................................................................................... 24 mg# Compositions ...................................................................................................... 26 Multi-element Spider Diagrams.................................................................................. 27 REE Diagrams ............................................................................................................ 28 Tectonic Discrimination Diagrams ............................................................................. 29 Sulfide Mineralization........................................................................................................ 30 Sulfide Saturation............................................................................................................... 30 Sulfide Type....................................................................................................................... 32 PGE+Au Grade .................................................................................................................. 33 PGE Depletion ................................................................................................................... 33 Exploration Recommendations .......................................................................................... 34 Contact-style Mineralization .............................................................................................. 37 PGE reef Mineralization..................................................................................................... 38 Field and Geochemical Study of the Deer Lake and Newton Lake Complexes........................ 40 Diabase Dikes............................................................................................................................ 41 CONCLUSIONS .......................................................................................................................... 41 REFERENCES CITED AND BIBLIOGRAPHY ............................................................................ 42

p.2


EXECUTIVE SUMMARY

This report summarizes a study of the potential for platinum group elements (PGE) and associated metals in mafic to ultramafic intrusions in Minnesota, excluding the Duluth Complex. Much of the interest in these intrusions follows discoveries during the last decade of PGE deposits in adjacent parts of Canada. It is the intent of this study to guide exploration, resource management, and land-use planning efforts by evaluating the potential for PGE in geologically similar terranes in Minnesota. The intrusions range in age from Mesoproterozoic (1.1 billion years old) to Archean (more than 2.6 billion years old), and are exposed in outcrop and occur in the subsurface in nearly all regions of the state. The report contains an extensive database of geologic, geochemical, geophysical, and geochronologic information about more than 200 such intrusions. It was compiled from publications and unpublished exploration company file data, and augmented by new field mapping, drill core logging, petrographic studies, and geochemical analyses. In addition, the report provides detailed assessments of select intrusions that are inferred to have some potential for PGE and are broadly representative of the diversity of intrusion types and ages. The study relies heavily on existing drill core archived in public repositories. This drilling was conducted to address a variety of exploration and scientific objectives, and rarely were those objectives specifically to assess PGE potential. Despite this, evaluation of geochemical data using the mineralization parameters of sulfide saturation, dominant sulfide mineralogy, and PGE depletion, combined with what can be determined about the igneous stratigraphy of each intrusion, allow us to make predictions with varying degrees of certainty about whether and where significant PGE mineralization might be found within particular intrusions. We conclude that nearly all types of intrusions in this study have some PGE potential; however, ultramafic (komatiitic) and mafic (tholeiitic) intrusions that display some petrographic evidence of differentiation are the most promising. The alkalic rocks, diabase dikes, and intrusions having mixed magmatic compositions are considered to have less potential. Of the 38 intrusions selected for detailed analysis, about 20 intrusions in various regions of the state display geochemical attributes indicating significant potential for PGE.

p.3


INTRODUCTION This report describes the results of a 4-year study of the potential for platinum group

elements (PGE) and associated metals in mafic to ultramafic intrusions in Minnesota, excluding the Duluth Complex. The study was funded during parts of 2 biennia by the Minerals Diversification program of the State Legislature on recommendation of the Minnesota Minerals Coordinating Committee, and the State Special appropriation to the Minnesota Geological Survey. Phase 1 of the study, conducted during the 2002-2003 biennium, and produced an extensive database of geologic, geochemical, and geochronologic information about the intrusions, which was released to open-file (Jirsa and others, 2003b). Phase 2 of the study, conducted 2004-2006, refined the earlier data and focused effort on select intrusions identified as having some potential for PGE and being broadly representative of the diverse intrusion types and ages. Geochemical data from Phase 2 were released to open-file in 2006 (Jirsa and others, 2006a; MGS website: geo.umn.edu/mgs/released documents). This Final Report summarizes the work conducted and results obtained during both phases of the study.

Rationale

Exploration in Minnesota for polymetallic deposits including PGE has traditionally focused on mafic intrusions of the Mesoproterozoic (approximately 1.1 Ga) Duluth Complex for obvious reasons—the complex contains known occurrences of PGE associated with Cu-Ni deposits that were discovered decades ago (summarized in Severson and others, 2002). Three of these deposits are currently in the environmental review and permitting process pursuant to mining. By contrast, this study investigates the many smaller intrusions and layered mafic complexes scattered throughout the state in rocks ranging in age from Paleoarchean to Mesoproterozoic. Much of the interest in these varied intrusions follows discoveries during the last decade of PGE deposits in correlative terranes in Canada. Some of the Minnesota intrusions appear lithologically similar to the so-called “Quetico” and “Atikokan” intrusions that are the targets of PGE exploration in Canada (eg., MacTavish, 1999; Vaillancourt and others, 2001). The most productive Canadian example is the Lac des Iles PGE-Cu-Ni Mine that lies within a suite of such intrusions (Brugmann and others, 1989). Some exploration work by private companies is already underway in several Minnesota intrusions. This project characterizes the various types of mafic intrusions in Minnesota, in an effort to assess the state’s resource potential and encourage further exploration for PGE in the diverse geologic terranes throughout the state.

p.4


Objectives and Methods Phase 1 of the study entailed compiling both written and GIS-based inventories of mafic,

ultramafic, and alkalic intrusions using all available outcrop, drill hole, and geophysical information (Jirsa and others, 2003b). In addition, a small number of these intrusions were dated using 40Ar/39Ar analyses of magmatic hornblende and biotite separates, which contributed to establishing temporal constraints for mineral deposit models. Many of the intrusions were discovered during geologic mapping by the Minnesota Geological Survey and were variously described in published maps, information circulars, and reports of investigations cited in the references. Others were discovered by exploration companies in search of base-metal or other mineral deposits, including intrusions represented by a set of drill cores acquired in an apparent search for diamonds. These varied sources of information provide a general understanding of the attributes of such intrusions, but little insight pertinent to their potential for PGE. In fact, prior to this endeavor, few of the intrusions had been thoroughly described and analyzed, and almost none had been analyzed for content of PGE or other geochemical indicators useful for predicting large concentrations of these elements. This project was designed to compile all pertinent data regarding the various intrusions, and investigate their geologic setting, composition, and potential for occurrence of PGE mineralization. It should be noted that many of the intrusions are represented by a single or a very few drill cores—a condition that limits our understanding of the dynamics of individual magmatic systems. Any inferences made by the authors about magmatic phases, differentiation, intrusion geometry, and potential metallic mineral content must be considered in this light. GEOLOGIC SETTING

Figure 1 and the associated ARCMAP PROJECT show the distribution and geologic context of more than 180 major mafic intrusions and intrusive complexes exposed in outcrop or intersected in drill cores in Minnesota. The intrusions are inferred to range in age from Archean to Mesoproterozoic. This is based on emplacement relationships deduced from a small number of high-precision U-Pb dates of intrusions and host rocks, field work, geophysical signatures, and new Ar-Ar analyses of 7 intrusions. In addition, we identify 320 speculative intrusions and the trajectories of thousands of diabasic dikes from geophysical anomalies (Figs. 2 and 3). The inventory of intrusions was augmented during Phase 2 of the study by conducting core logging, petrographic study, geochemical analyses, and geophysical modeling of 38 select intrusions (Fig. 4 and INTRUSION_ATTRIBUTES.xls). Although some intrusions are locally exposed at the land surface, most are covered by Phanerozoic strata, including Paleozoic and Mesozoic sedimentary rocks and Cenozoic glaciogenic sediments. The thickness of the latter is shown on Figure 5.

p.5


Figure 1—Generalized geologic terrane map of Minnesota (modified from Morey and

Meints, 2000) showing the distribution and geological association of various types of mafic intrusions and drill holes that intersected small intrusions. Intrusions depicted are mafic to ultramafic, and those locally containing mafic phases. County outlines are shown.

p.6


Figure 2—First vertical derivative map of aeromagnetic data showing major faults (white,

solid lines) and outlines of geologic terranes (white, dashed and solid lines).

p.7


Figure 3— Generalized geologic terrane outline map showing the distribution of known and

speculated intrusions; the latter inferred from geophysical anomalies. Note that diabase dikes are shown locally beneath younger bedrock cover, as projected from geophysical data.

p.8


Figure 4—Generalized geologic terrane map showing the distribution of select mafic

intrusions investigated in detail (see INTRUSION_ATTRIBUTES.xls). County outlines shown in gray.

p.9


Figure 5—Depth to bedrock map of Minnesota (from Lively and others, 2006) showing

major faults and terrane boundaries. Note that bedrock includes Phanerozoic strata.

p.10


PROJECT RESULTS Inventory of Intrusions

The INVENTORY06.pdf file contains data from more than 180 individual intrusions that are either exposed in outcrop or are represented by archived drill core. The project work conducted on each intrusion is in part dependent on the amount and quality of information available, but included some combination of geophysical delineation, outcrop mapping, drill core examination, petrography, geochemical analysis, acquisition of geophysical rock properties, and geophysical modeling. The inventory provides details of location, lithology, geologic setting, types of work conducted on each intrusion, and references to additional data. Detailed chemostratigraphic study of the well-exposed Deer Lake and Newton Lake layered mafic sill complexes was conducted by the Natural Resources Research Institute as a separate, but related objective, and is summarized below.

Geochemical Database

Excluding data from the Deer Lake Complex, a total of 679 geochemical analyses were compiled for this study from over 71 mafic and ultramafic intrusions. These analyses are listed in the file GEOCHEMICAL_DATA.xls. This file contains two worksheets: Worksheet GDT is the main geochemical data table and Worksheet CIPW tabulates CIPW normative calculations determined from major and minor element analyses.

Worksheet GDT

The background information provided on the main geochemical data table includes:

Sample – lists the sample label reported from the various data sources (see below)

DDH/OC – name of the drill core or outcrop sample. Individual drill core or outcrop samples can be located using the ArcMap project. Samples from outcrops and drill cores state-wide are shown on themes ocsam.shp and intrusions_pt.shp.

Depth (ft) –mean depth in feet below surface for samples taken from drill core.

Interval – depth interval, in feet below surface, for samples taken from drill core; samples less than 2’ in length (most samples collect for this study) are indicated by listing the sample interval at “1’” and centered on the depth indicated in the preceding column.

Intrusion – name of the intrusions cited in this report and described in the intrusion inventory (INVENTORY06.pdf).

Intrusion Type –general intrusion type as defined in Table 1. Samples of wall rock adjacent to intrusions or country rock inclusions are also noted.

Normative Rock Type – rock names applied to samples with CIPW normative compositions calculated from major and minor element analyses (see Worksheet CIPW) that are determined

p.11


by the criteria set out in Table 3. Rock names in boldface correspond to those determined from CIPW calculations of the average compositions also shown in boldface.

Sample Descriptions – lithologic descriptions based on core logs and petrographic observations. Drill core sample descriptions based on petrographic observations are italicized and indicate the sample depth in parentheses.

Comments – miscellaneous information on the location of samples relative to the intrusion contacts or other distinctive features represented by the sample.

Data Source – the original source of the geochemical analyses reported here come from three basic sources:

• geochemical and assay data contained exploration company reports on file with the Minnesota Department of Natural Resources, Lands and Minerals office in Hibbing; this is indicated by “DNR (company name - DNR folder number)”

• geochemical analyses from published reports by the Minnesota Geological Survey, the MN Department of Natural Resources, the U.S. Geological Survey, and the Natural Resources Research Institute; reference citations for the various reports are listed at the bottom of the Data Source column

• whole-rock and PGE analyses acquired for the two phases of this study; these analyses are highlighted in yellow.

Analyst – the geochemical lab where the analysis was obtained. Three different commercial labs were used for the analyses conducted for different phases of this study. Phase 1 analyses were acquired with the 4Eexpl + 1Cres analytical packages from X-Ray Laboratories of Don Mills, Ontario. Phase 2 analyses were acquired with two different analytical packages from Actlabs of Tucson, Arizona (4litho + 4B1+1C-exp2 and 1H+1C-exp2) and with the 3B-MS+1T-MS analytical packages from Acme Laboratories of Vancouver, British Columbia.

Analytical Methods – the types of analyses used to acquire the geochemical data; methods include:

FA - fire assay XRF - X-ray fluorescence analysis ICP - induced coupled plasma analysis ICP-MS - ICP w/ mass spectrometer finish INAA - induced neutron activation analysis AA – atomic absorption analysis

Major and minor element analyses reported in columns M through X were conducted by two general methods. X-ray fluorescence spectrometric analysis (XRF) produces accurate analyses of all major and minor elements, as well as a loss on ignition determination (LOI), which are reported in oxide weight percents. A number of analyses conducted for this study acquired major and minor element analyses by induced coupled plasma analysis with a mass spectrometric finish (ICP-MS). However, these analyses do not analyze for SiO2 and the data are reported in atomic percentages. This atomic data is reported on the far right columns of the spreadsheet (columns

p.12


BU to CC) and are converted to oxide weight percents by using the conversion factors listed in columns M-W of row 4. A crude estimate of SiO2 abundances are determined by subtracting the converted oxide weigh percents from 100%. Because the analyses lack an estimate of LOI, these SiO2 estimates are clearly too high. In addition, it was discovered that ICP-MS analyses conducted by Actlabs grossly underreported Al, probably due to incomplete dissolution. Theses inaccurate analyses of SiO2 and Al2O3 are highlighted in italics and will be discussed below in the section on CIPW normative calculations. For intrusions with three or more analyzed samples, average major and minor element compositions were determined (shown in boldface). Sample analyses included in determining the averages are shown in blue. Lastly, whole rock mg#’s (=MgO/(MgO+FeO+Fe2O3); mole %) are calculated in column Z.

Various combinations of trace element abundances are listed in columns AA through BT. All data are reported in parts per million (ppm) except for PGE (Pt., Pd, and Au) assays, which are given in parts per billion (ppb), and sulfur, which is reported in weight %. Abundance reported to be less than detection limits are indicated by a < symbol.

Worksheet CIPW

Whole rock data and some background information for samples listed in Worksheet GDT with major and minor element analyses are compiled in this worksheet for the purpose of calculating the CIPW normative composition. Such calculations are a useful way to estimate the original igneous mineralogy of the many altered and metamorphosed samples collected in this study. The major and minor element data were run through the CIPW normative calculation program that is part of the IGPET99 software package (Carr, 1999). The resultant mineral components are reported in mole percentages normalized to 100%. These data are used to determine normative rock names by the criteria given in Table 3 and using the normative mineral compositions and other compositional parameters calculated from them (see columns AS-BD).

As noted above, analyses conducted by ICP-MS did not measure for Si and one set of Phase 2 data from Actlabs underreported concentrations of Al. Yet these two major elements are critical to reasonably accurate CIPW norm calculations. Comparison of these Al data with XRF analyses of similar rock types from the same core imply that the Actlab Al2O3 analyses are routinely off by about 7 wt%. Therefore, this amount was added to all Actlab ICP-MS analyses (34 analyses from 9 intrusions; compare data in worksheets MGT and CIPW). SiO2 concentrations for all ICP-MS analyses (65 analyses from 16 intrusions) were estimated by subtracting from 100% the weight percent concentrations of all other major and minor oxide abundances and the average loss on ignition abundance for all other similar samples for that core (if other LOI data are not available, LOI is assumed to be 1%). The re-estimated values for Al2O3 and SiO2 are indicated in red italics in worksheet CIPW. The normative compositions calculated from these modified analyses should be considered rough approximations.

Geochronologic Study

p.13


The age of some intrusions described in this report can be generally constrained by their relationship to regional tectonic fabrics in host rocks; however, the absolute age of many of the post-tectonic intrusions is poorly known. The dearth of zircon in many of the mafic intrusions precludes the use of precise U-Pb dating, and less precise methods were necessary. To that end, several plutons were dated using the CO2 laser 39Ar/Ar40 incremental heating technique at the University of Wisconsin-Madison Rare Gas Geochronology Laboratory (Keatts and others, 2003). For late-stage (post-tectonic), shallowly emplaced plutons containing primary magmatic hornblende, Ar/Ar mineral ages are likely to approximate the crystallization age. In regions with a more protracted thermal history (i.e., low-grade metamorphism, slow-cooling, etc.), the Ar/Ar data provide minimum ages for the mafic plutons. The mafic intrusions selected for this study represent a broad range of geologic settings, including 1) small mafic plutons emplaced into Paleoproterozoic supracrustal and intrusive rocks within the Penokean orogen (samples 264, R17); and 2) varied, primarily late- to post-tectonic intrusions in supracrustal rocks of the Archean Wabigoon (samples A1, B21, UBD) and Wawa (samples K15, LP, ANA) subprovinces of Superior Province. The Ar/Ar results from eight separate intrusions are reported here. East-central Minnesota. A hornblende grain (R17) from a sample of medium-grained hornblendite from the Tibbett's Brook intrusion cutting the East-central Minnesota batholith in Morrison Co. yields a plateau date of 1.770 ± 0.006 Ga from 4 contiguous increments constituting 74% of the gas released. A biotite grain (264) from a sample of coarse-grained biotitic olivine gabbronorite cutting the Little Falls Formation in Morrison Co. yields a plateau date of 1.791 ± 0.008 Ga from 5 contiguous increments constituting 68% of the gas released. Wawa Subprovince. A hornblende grain (ANA) from a sample of prismatic hornblende diorite collected near Red Lake in Beltrami Co. yields a near-plateau date of 2.587 ±0.012 Ga in 5 non-contiguous increments constituting 50% of the gas. A biotite grain (K15) from a sample of biotite granodiorite porphyry collected in Norman Co. yields a plateau date of 2.639 ± 0.007 Ga from 6 contiguous increments constituting 79% of the gas released. A biotite grain (LP) from a sample of porphyritic syenite collected at the Wawa-Quetico subprovince boundary in St. Louis Co., in the Linden Pluton, yields a plateau date of 2.666 ± 0.006 Ga from 7 contiguous increments constituting 88% of the gas released. Wabigoon Subprovince. A hornblende grain (B21) from the Oaks intrusion leucodiorite sampled near the Vermilion Fault in Roseau Co. yields a plateau date of 2.671 ± 0.008 Ga from 8 contiguous increments constituting 75% of the total gas released. A hornblende grain (A1) from the Black River gabbro, collected in Roseau Co., yields a plateau date of 2.685 ± 0.011 Ga from 11 contiguous increments constituting 90% of the total gas released. A hornblende (UBD) from a sample of hornblende-biotite gabbro collected in Koochiching Co. north of the Rainy Lake-Seine River Fault yields a plateau date of 2.695 ± 0.007 Ga from 6 contiguous increments constituting 49% of the gas released.

The new mineral age data are significant, in that previous estimates of ages relied solely on the fact that the intrusions post-date metamorphism and deformation in each terrane in which they occur, and in many places the timing of those events is not well constrained. The data from mafic

p.14


plutons in the Wabigoon subprovince are synchronous with the last deformation event (D2) dated in the range 2.685-2.674 Ga. Mafic plutons from the Wawa subprovince give an 80 m.y. age range from 2.58 to 2.66 Ga. Interestingly, the Linden Pluton gives a biotite date concordant (within error) with the youngest mafic pluton from the Wabigoon subprovince. The younger spread of ages from the Wawa subprovince is consistent with southward growth of the Superior Province during the latest Archean. Mafic plugs evident from aeromagnetic maps in east-central Minnesota (unit PP on Table 1) could previously be constrained only by the age of their youngest host rocks—approximately 1.8 Ga. Thus, their age might have been as young as Mesoproterozoic, or even younger. The new data indicate that the mafic plugs are comagmatic with the circa 1.775 Ga east-central Minnesota batholith (Holm and others, 2005).

One additional age date has been acquired for the "BO-1" intrusion in Fillmore County, southeastern Minnesota. The intrusion lies beneath as much as 1000 feet of Paleozoic sedimentary rocks, and within a magnetically complicated zone east of the Mesoproterozoic Midcontinent rift. The U-Pb zircon date of 1760+/-9 Ma (Van Schmus, personal communication, 2006) temporally places the intrusion in the Paleoproterozoic Yavapai terrane (Cannon and others, 2005). Geophysical Data

The project created various geophysical images and grids (state-wide grids included on the associated ARCMAP PROJECT), and permitted acquisition of a small amount of new, site-specific gravity data and related geophysical modeling. Most of the intrusions are magnetic or contain magnetic phases, making them amenable to delineation on aeromagnetic maps at varied scales (Figure 2 and ARCMAP PROJECT). Derivative gravity maps and gravity profiles were used locally to refine intrusion geometry and identify phase boundaries in areas where the geology is moderately well-constrained by outcrops and drill holes—specifically the Warroad intrusion in the Wabigoon subprovince, and the Gheen pluton in the Quetico subprovince. The two intrusions represent widely divergent types of PGE targets; the Gheen intrusion consists of a small, irregular mass of dike-like intrusions that are very heterogeneous, with compositions ranging from pyroxenite to syenite. In contrast, the Warroad intrusion is thought to be a body of gabbroic rocks that is 1-3 miles wide and at least 12 miles long. The work included acquisition of detailed gravity data along three profiles across the Gheen intrusion, and four profiles across the Warroad intrusion. The combination of new and pre-existing gravity data result in station spacings along most of the survey profiles ranging from 0.25 to 1.0 miles.

The three profiles in the Gheen area indicate that the intrusion is associated with a positive gravity signature of 1-3 milligals amplitude, implying that the overall composition of the intrusion is mafic. Scattered bedrock exposures indicate that mafic rocks occur in about equal proportion to rocks of more intermediate composition. Local extension of the gravity high beyond the presently mapped contacts may correspond to places where mafic intrusions were emplaced into wall rocks. Localized negative anomalies of 0.5 to 1.0 milligals amplitude that are

p.15


superimposed on the positive signature of the intrusion probably reflect small troughs on the bedrock surface filled with glacial deposits.

The Warroad intrusion lies at depths of 100-200 feet beneath glacial sediments, based on several drill holes that intersected the intrusion. The four profiles crossing the Warroad intrusion indicate a strongly positive anomaly that has an amplitude of 18-34 milligals. Model studies along the 4 profiles indicate that the observed gravity signature can be produced by a series of vertical, 5 km-deep slabs that have densities ranging from 2.70 to 2.75 gm/cm for undisturbed wall-rock, and 2.79 to 3.02 gm/cm for the intrusion and associated contact zones. Within the intrusion, bodies having densities greater that 2.90 gm/cm are interpreted to represent gabbroic rocks, whereas bodies with densities ranging from 2.79 to 2.87 gm/cm are tentatively interpreted to represent zones containing a significant proportion of relatively less dense wall rock, either in the form of large inclusions or pendants. Zones of intermediate (2.79-2.86 gm/cm) density that extend beyond the intrusion contacts, as presently mapped, are interpreted to represent wall rock that has been either contact metamorphosed, or intruded by thin bodies of mafic rocks.

The results of this study—though limited in scope—demonstrate that the gravity method can readily provide useful information regarding the composition and structure of intrusions, and therefore the method could be a useful supplement to some platinum exploration programs. The gravity profile data are on-file with the Minnesota Geological Survey, and details are available on request.

Attributes of Select Intrusions During phase 2 of the study, particular intrusions were selected for detailed work that

included core logging, petrographic study, and geochemical analyses. The results are compiled in the file INTRUSION_ATTRIBUTES.xls. The file is composed of two worksheets - worksheet IAT is the main intrusion attribute table and worksheet DDHA gives information on the diamond drill holes cited in the attribute table. The information provided in worksheet DDHA includes TRS locations of the drill holes, core recovery depths and intervals over which mafic or ultramafic intrusive rocks were encountered, when the core was drilled and by whom, and where the core is stored in the MNDNR library.

The intrusion attribute table is intended to serve as a brief summary of the salient lithologic, geochemical and PGE-related mineralization features of 38 different mafic and ultramafic intrusions distributed throughout Minnesota, outside the Duluth Complex, and related intrusions of northeastern Minnesota. The 38 intrusions included in the intrusions attribute table were selected from the 180 listed in the intrusion inventory (INVENTORY06.pdf) because they have significant intervals sampled by drill core, multiple geochemical analyses (including assay data) from various depths in the intrusion (see GEOCHEMICAL_DATA.xls/ worksheet GDT), and because they provide a broad geographic and geologic representation of different intrusion types from across the state. Diabase dikes were excluded from this list due to their prevalence and generally poor PGE potential. Some intrusions sampled by short cores or with only one analysis

p.16


are also included to provide a few representative examples of intrusions from provinces in central and southern Minnesota. The criteria used to define and distinguish the various intrusion attributes reported in worksheet IAT (table columns) are discussed below. Intrusion Name

The informal names given to the various intrusions or groups of intrusions are typically based on geographic locations near the drill hole site(s) or main outcroppings. Two locations—Oaks, and Warroad—host several closely-spaced intrusions, and thus share common geographical names, but are distinguished in each location by the directional modifiers (e.g., Oaks West, Warroad South). Some intrusions are named after the defining drill hole (e.g., UBD-3, ANA, OB-329), and some after the county in which they occur (e.g., Fillmore B-1, Meeker BKV). Some intrusions are named after larger felsic to intermediate intrusive complexes with which they may be associated (e.g. Grygla, Mentor, Gheen, Linden).

County

The county (or counties) in which the intrusion occurs is given as a general guide to the approximate statewide location of the intrusion. DDH or OC

Most of the intrusions on this list are based on intersections in one or more drill cores stored at the MNDNR core library in Hibbing (see worksheet DDHA). The drill core labels are those listed on the DNR’s drill hole register. The depth interval (in feet) which intersects the intrusion is given in the parentheses after the drill core label. Those intrusions based on outcrop samples are indicated by “(oc)” after the sample label. A couple of samples from the Mentor intrusion (M92-1, M92-6) were collected from a waste pile excavated from an underground gas storage cavern (see MGS IC-40) and are labeled as “float”. Province

The major geotectonic province or subprovince hosting each intrusion is listed in this column. Figure 4 shows the distribution of these intrusions within the major provinces/subprovinces In Minnesota. A general description of the geology and tectonic elements of these provinces is given by Southwick (1996), Morey and Southwick (1991), Southwick and Chandler (1996), and Holm and others (2005). Intrusion Type

This column gives an interpretation of the basic intrusion type based mainly on intrusion size and shape, general lithologic characteristics, internal structures, stratigraphic variability in lithology and geochemistry, and degree of metamorphism/tectonism. These criteria are defined in Table 1 below.

p.17


Table 1. General characteristics of intrusion types recognized in this study.

Intrusion Type

General Characteristics

Layered ultramafic-mafic intrusion

Large, well differentiated intrusion containing both ultramafic and mafic cumulates; commonly display meter-scale modal layering

Ultramafic intrusion

Composed of melanocratic rocks dominated by orthopyroxene, clinopyroxene, and/or olivine (and their alteration products); commonly displays cumulus texture and evidence of differentiation

Alkaline mafic intrusion

Composed of quartz-poor, mesocratic to melanocratic rocks with significant amounts (>15%) of primary hornblende, biotite and/or alkali feldspar; may display cumulus texture and evidence of differentiation

Tholeiitic mafic intrusion

Composed of mesocratic rocks dominated by plagioclase and pyroxene, with variable amounts of olivine and Fe-Ti oxide, and minor (<5%) amounts of hornblende, biotite, apatite, and quartz; commonly display cumulus texture and evidence of differentiation

Anorthositic mafic intrusion

Dominated by leucocratic cumulates containing >70% cumulus plagioclase

Intermediate intrusion

Intrusions composed of variable amounts of plagioclase, quartz and/or alkali feldspar and 15-50% mafic mineral phases (dominantly hornblende, but may also include biotite, pyroxene, and Fe-Ti oxide)

Diabase dike

Fine-grained at margins to coarse-grained in centers, non-cumulate, mineralogically homogeneous, mafic intrusion (typically < 100 m wide) that cross-cuts country rock structure

Hypabyssal mafic sill

Generally fine-grained, non-cumulate, (typically < 100 m thick), homogeneous mafic intrusion that is grossly conformable to country rock structure

Lamprophyric intrusion

Alkaline melanocratic mafic intrusion that displays porphyritic texture of mafic phases

Mafic intrusive breccia

Mafic to intermediate rock rich in cognate xenoliths

Metagabbroic intrusion

Mafic intrusions that have been strongly recrystallized by greenschist to amphibolite grade metamorphism such that original igneous mineralogy and texture are largely obscured

Tectonized mafic intrusion

Mafic intrusions that have been strongly sheared, ductily deformed, and variably metamorphosed such that original igneous mineralogy and texture are largely obscured

Mafic-felsic migmatite

Complex multiphase mixtures of coarse-grained mafic to felsic rock types; typically composed of mafic enclaves within intermediate to felsic rock types and commonly showing evidence of metamorphic recrystallization and ductile deformation

p.18


It should be stressed that the interpretation of intrusion type is typically based on limited data.

Except for the Gheen intrusion and some lamprophyre intrusions in the Quetico Subprovince, all other intrusions are known only from drill core. Moreover, except for thin sills, most cores do not completely section the intrusions. Some cores intersect one contact, whereas others are completely contained within the intrusion.

Mafic, alkaline, and ultramafic intrusions that display cumulate textures and have evidence of internal differentiation can be parts of intrusions that are on the order of several kilometers in thickness. Therefore, with the longest core being just over 150 meters (WF-4 - 160m, 533’), it is clear that many large, well-differentiated intrusions may be difficult to recognize. Layered mafic-ultramafic intrusions, which are typically Archean and host most of the world’s known PGE deposits (e.g., Bushveld, Stillwater, Great Dyke, Fennoscandian intrusions), would be particularly hard to identify from the limited drill core available for Minnesota intrusions. However, most of these massive intrusions are typically associated with significant geophysical anomalies. The only Archean intrusions that might be considered layered ultramafic-mafic intrusions based on the presence of cumulus textures, a thickness exceeding 1 kilometer, and their association with sizeable geophysical anomalies, are the Grygla and Mentor intrusions. The only Paleoproterozoic intrusions that have these characteristics are the deeply buried Fillmore B-1 and BO-1 intrusions of southeastern Minnesota. Age

Before the Ar-Ar age dating conducted by Daniel Holm and his colleagues for this project (as reported in the Geochronologic Study above), no precise radiometric dates of mafic intrusive rocks in Minnesota had been acquired. The six Ar-Ar dates on amphiboles reported by Keatts, Holm, and Jirsa (2003) are listed in this column. One other unpublished date is a U-Pb age of zircons from an apatitic oxide gabbro from the Fillmore BO-1 intrusion (Van Schmus, 2006; Holm and others, in prep.).

For the other intrusions, their general ages or possible range of ages are estimated from their geologic setting and metamorphic condition, and are given in terms of Precambrian eras or periods (Table 2) as defined by Plumb (1991). All mafic and ultramafic intrusions listed in this table and occurring in the Wabigoon, Quetico, and Wawa subprovinces of the Superior provinces are thought to be Neoarchean in age (2800-2500 Ma). No igneous rocks older than this have been found in these subprovinces in Minnesota, and the only intrusions younger in age are the Kenora-Kabetogama dike swarm at about 2070 Ma (Schmitz and others, 2006).

All of the intrusions emplaced in the Penokean orogenic province are post-deformational (<1820 Ma; Holm and others, 1988; 2005) and are thought to be late Paleoproterozoic (1800-1600; Stratherian period) or younger. The Tamarack intrusion, in particular, has similarities to ultramafic intrusions of the 1.1 Ga Midcontinent Rift. The others appear to be contemporaneous with the emplacement of the East-central Minnesota batholith (~1760 Ma, Holm and others, 2005).

p.19


The ages of intrusions emplaced into the 3+ Ga Minnesota River Valley gneiss terrane are difficult to constrain. The last major intrusive and metamorphic event occurred during the Neoarchean (~2.6 Ga) with the accretion of the various subterranes, creating the greater Superior province (Southwick and Chandler, 1996; Schmitz and others, 2006). Most of the intrusions are unmetamorphosed and, therefore, are likely Neoarchean (post-tectonic) or younger. The high metamorphic grade of the Cottonwood intrusion implies that it is Neoarchean (syntectonic) or older; perhaps as old as 3.0 Ga. The low grade metamorphism of the Adrian intrusion, which occurs at the southern margin of the MRV or possibly in a Paleoproterozoic terrane, could be Neoarchean (2.8-2.5 Ga) to Mazatzal (~1.65 Ga) in age.

Prior to the 1760+/-9 Ma, U-Pb age acquired by Van Schmus (2006) on a sample from Fillmore BO-1, both Fillmore County intrusions were thought to be Mesoproterozoic in age. The complex aeromagnetic patterns adjacent to the Keweenawan Midcontinent Rift are similar to the geophysical patterns over the Duluth Complex, which flanks the main axis of the rift in the Lake Superior region. Although lithologic similarities of the two Fillmore County intrusions (both oxide gabbro cumulates) imply that they may be of similar age, Fillmore B-1 is distinctly unaltered compared to the strongly altered Fillmore BO-1. Therefore, it is possible that Fillmore B-1 could be Mesoproterozoic. Table 2. Precambrian eras (boldface) and periods (after Plumb, 1991) and major geologic events

in the Lake Superior region (Southwick, 1996; Southwick and Chandler, 1996; Holm and others, 2005, Schmitz and others, 2006)

Age Age Interval Local tectonic eventsPaleoarchean 3600-3200 Ma Morton/Montevideo event?? Mesoarchean 3200-2800 Ma MRV thermal event?? Neoarchean 2800-2500 Ma Kenoran orogeny Paleoproterozoic 2500-1600 Ma Sideran 2500-2300 Ma Rhyacian 2300-2050 Ma Southern Superior Prov. rifting Orosian 2050-1800 Ma Penokean orogeny Statherian 1800-1600 Ma Yavapi/Mazatzal orogenies Mesoproterozoic 1600-1000 Ma Calymmian 1600-1350 Ma Wolf River batholith Ectasian 1350-1200 Ma Stenian 1200-1000 Ma Midcontinent rift

Geologic Setting

The host rocks into which the intrusions were emplaced are generally described here. The geologic setting is determined, where possible, from direct drill core evidence. In those cases where direct evidence is lacking, the geologic setting is inferred from regional geologic maps,

p.20


which are themselves based largely on interpretations of aeromagnetic and gravity data integrated with scant outcrop and drill core information. For intrusions in the southern part of the state, the geologic setting is typically only generally known. The geologic maps which pertain to the geologic setting are listed in the reference column. General Lithologic Description

Lithologic attributes given in this column are based on core logs and petrographic observations conducted for this study. Intrusions based on core logs and petrographic descriptions taken from other reports are indicated. Descriptions based on core logs alone are indicated by “No TS”. The information presented includes primary and secondary mineralogy and texture, stratigraphic variations in rock type, description of internal structures, and the nature of intrusive contacts, if intersected in drill core. Igneous rock names generally adhere to the classification of Streckeisen (1976). Textural terminology used is defined in Miller and others (2002). Normative Rock Types

Because alteration or metamorphism typically obscures the original igneous mineralogy in many of the intrusions studied here, an attempt is made to estimate the original mineralogy by calculating the CIPW normative compositions of all samples that have been analyzed for major and minor element concentrations. A total of 290 analyses from 68 different intrusions were applied to the CIPW norm calculation program in IGPET 99 (Carr, 1999) and the results are summarized in the geochemical data table (GEOCHEMICAL_DATA.xls/worksheet CIPW).

As reported in the section on the geochemical database, major and minor element concentrations determined ICP-MS did not include SiO2 in the analytical package and, for the analyses conducted by Actlabs, it was discovered that Al2O3 concentrations were underestimated (similar analyses by Acme Analytical seem to be reasonably accurate). These data are denoted by italics in GEOCHEMICAL_DATA.xls/worksheet MGT. Because CIPW normative calculations are critically dependent on SiO2 and Al2O3 abundances, some reasonable estimates of these concentrations were necessary to calculate the CIPW norms analyses, as described in the geochemistry section. Again, the normative compositions calculated from these modified analyses should be considered rough approximations.

From the calculated normative mineral components listed in GEOCHEMICAL_DATA.xls/ worksheet CIPW, rock names were assigned to each analysis on the basis of the criteria listed in Table 3. These rock classifications are modified from those of Thompson (1984) for mafic and ultramafic rocks and Streckeisen and Le Maitre (1979) for intermediate to felsic rocks. The compositional components used to define the different rock names are:

CI – color index (Ol+Di+Hy+Mt+Ilm) Q’ – Q/(Q+Or+Ab+An) F’ - (Ne+Lc+Kp)/(Ne+Lc+Kp+Or+Ab+An) Feld – An+Ab+Or

p.21


Foid - (Ne+Lc+Kp) ANOR – An/(An+Or) AN – An/(An+Ab) mg# - MgO/(MgO+FeO), mole%

The normative rock names determined for each analysis by these classification criteria are listed in INTRUSION ATTRIBUTES.xls/worksheet IAT and GEOCHEMICAL_DATA.xls/worksheets CIPW and GDT in the columns labeled “Normative Rock Types”. The various normative rock types determined for each intrusion are then compiled in worksheet IAT from most to least common.

Table 3. Criteria defining normative rock types.

Normative Rock Group

Group Attrib.

Normative Rock Subgroups

Subgroup Attributes

Mineral Modifier Attributes

ULTRAMAFIC

AN>40 CI>75

DuniteLherzolitePeridotite

Pyroxenite

Ol/(Di+Hy)>5 Ol/(Di+Hy) 1-5 Ol/(Di+Hy) 1-0.2 Ol/(Di+Hy)< 0.2

Oxide – (Mt+Il) >5 Foid – F’ >0

MAFIC

AN>40 CI 50-25

Troctolite#*Gabbro

#*Gabbronorite#Norite

Ol/(Di+Hy)>5 Hy/Di<2, Di>5 Hy/Di>2, Di>5 Hy/Di>2, Di<5, Ol<2

Leuco- – CI <25 Mela- – CI=75-50 *Olivine – Ol >2% #Quartz – Q >2% Oxide – (Mt+Il) >5 Apatitic – Ap >3 Foid – F’ >0

INTERMEDIATE

AN<40 CI>15

FerrodioriteFerromonzodiorite

FerromonzoniteFerrosyeniteFerrotonalite

Ferrogranodiorite

ANOR>75, Q’<20 ANOR=75-50, Q’<20 ANOR=50-25, Q’<20 ANOR<25, Q’<20 ANOR>60, Q’>20 ANOR=60-30, Q’>20

Olivine – Ol >2% Quartz – Q’=5-20 Nepheline–Ne >2% Oxide – (Mt+Il) >5 Apatitic – Ap >3

FELSIC

AN<40 CI<15

DioriteMonzodiorite

MonzoniteSyeniteTonalite

GranodioriteGranite

ANOR>75, Q’<20 ANOR=75-50, Q’<20 ANOR=50-25, Q’<20 ANOR<25, Q’<20 ANOR>60, Q’>20 ANOR=60-30, Q’>20 ANOR<30, Q’>20

Quartz – Q’=5-20 Nepheline–Ne >2%

p.22


Internal Differentiation

Compositional zoning within the intrusions that might indicate crystallization differentiation within the various intrusions is evaluated here. Specific zoning patterns that would be indicative of progressive differentiation driven by fractional crystallization include decreasing mg# and compatible element (CE) concentrations (e.g., Ni, Cr) and increasing incompatible element (IE) concentrations (e.g., K, P, Y, Zr, Ba, REE). Chilled margin effects commonly show opposite trends to differentiation (i.e., decreasing mg# and CE, increasing IE toward the margins).

About half of the intrusions are sufficiently well represented by samples and analyses to generally evaluate the nature of internal differentiation (see GEOCHEMICAL_DATA.xls/worksheet GDT). Many of those intrusions display compositional variations that are non-systematic or inconsistent. Metamorphism/Alteration

Assessment of the degree of hydrothermal/deuteric alteration or metamorphism that has affected each intrusion is based on 3 general criteria (Table 4) – maturity of recrystallization textures, preservation of original igneous texture and mineralogy, and secondary mineral assemblage. As defined in Table 4, evaluation of textural maturity is based on gradations in mineral habit from fibrous to more prismatic or poikilitic; changes in structural homogeneity or massiveness; and secondary minerals content; and grain size.

This was by far the most difficult attribute to evaluate for many intrusions in this study, largely because the identity of some mineral phases (particularly various types of amphibole) is difficult to determine conclusively in thin section. In addition, some sections have textures and mineral assemblages that do not fit easily into the criteria set out in Tables 4 and 4’. Consequently, some of the interpretations reported here should be treated as speculative.

Table 4. Criteria to evaluate intensity of alteration and metamorphism. Possible primary mineral phase replaced by secondary minerals is given in parentheses.

Alteration/ Metamorphism

Abbr

Recrystal-lization Texture

Preservation of Igneous Texture and Mineralogy

Secondary Mineralogy-Mafic

Secondary Mineralogy-Ultramafic

1 unmetamorphosed/ unaltered

UM

texture and mineralogy well-preserved

minor minor, esp. serpentine (ol)

2 weak hydrothermal/ deuteric alteration

WD

Low TM, non-foliated

texture well-preserved; weak to moderate replacement of igneous mineralogy

uralite (pyx) sericite/ sausserite (pl) serpentine/ iddingsite (ol) oxides (pyx/ol) leucoxene (ox)

serpentine/ iddingsite (ol)oxides (pyx/ol) uralite (pyx)

p.23


3 moderate-strong hydrothermal/ deuteric alteration

SD

Low TM non-foliated

texture largely preserved, moderate to complete replacement of igneous mineralogy

uralite/chlorite (pyx) sericite/ sausserite (pl) serpentine/ iddingsite (ol) oxides (pyx/ol) leucoxene (ox)

serpentine/ iddingsite (ol)oxides (pyx/ol) uralite (pyx)

4 greenschist facies metamorphism

GM Mod TM, possible foliation

texture largely preserved, complete replacement of igneous mineralogy

actinolite epidote chlorite/biotite plagioclase sericite/muscovite carbonate

actinolite serpentine talc carbonate oxide

5 amphibolite facies metamorphism

AM High TM, probable foliation

texture largely obscured, complete replacement of igneous mineralogy

hornblende epidote plagioclase biotite/phlogopite garnet

orthoamphibole tremolite

6 thermal metamorphism

TM granoblastic texture, nonfoliated

texture largely obscured, complete recrystallization of of igneous minerals

plagioclase pyroxene oxide

pyroxene

Table 4’– Criteria to define degrees of textural maturity (TM) Textural Maturity Mineral Habits Structural Homogeneity Grain Size Low fibrous matted microlites,

heterogeneous clusters fine

Moderate fibrous prismatic skeletal subpoikilitic

fibrous-prismatic masses, overgrowth coronas, anhedral massive grains

fine to coarse

High subprismatic, subhedral, porphyroblastic

anhedral to subhedral massive grains

medium to coarse

Geochemical Attributes

The salient geochemical attributes of the various intrusions are characterized by five different methods – 1) alkalic vs. tholeiitic vs. komatiitic discrimination plots; 2) mg#, 3) minor and trace element spider diagram normalization plots; 4) REE normalization plots; and 5) various tectonic environment discrimination diagrams. TIFF and WMF images of the various geochemical discrimination diagrams are compiled in subfolders within the GEOCHEMICAL PLOTS FOLDER. The complete geochemical database for these and other intrusions are compiled in GEOCHEMICAL_DATA.xls/worksheet GDT. Alk/Thol/CA/Kom (Alkalic/Tholeiitic/Calc-Alkalic/Komatiitic discrimination diagrams)- A basic classification of igneous rocks is between alkalic and subalkalic compositions. This is most commonly determined among volcanic rocks and noncumulate intrusive rocks by a total alkali vs. silica (TAS) diagram as shown in Figure 6. The intrusion data are plotted on TAS diagrams using

p.24


the IGPET99 program (Carr, 1999) and are compiled in the TAS subfolder. This diagram is inappropriate for intrusive rocks that are cumulates and thus are not representative of magma compositions. Intrusions that display cumulus textures are noted in the Alk/Thol/Kom classification column of the attribute table as “(Cumulate)”

Figure 6. Total Alkali-Silica diagram. Chemical classification boundaries for volcanic rocks

(grayscale lines and text) are taken from Le Maitre and others (1989) and the alkaline-subalkaline boundary curve (black dashed line) is taken from Irvine and Baragar (1971).

Intrusions that are classified as subalkalic can be further subdivided into tholeiitic, calc-alkalic, and komatiitic rock series by use of a Jensen cation plot (Figure 7). This diagram plots the cation proportions of Mg, Fe (total)+Ti, and Al and again is only applicable to volcanic rocks or noncumulate intrusive rocks. The intrusion data were plotted on ternary diagrams with these components using the IGPET99 program (Carr, 1999) and are compiled in the “Jensen” subfolder. A comparable diagram to the Jensen cation plot is the classic AFM diagram of Irvine and Baragar (1971), however, these plots do not discriminate a komatiite field. Nevertheless AFM diagrams for the various intrusions are plotted using IGPET99 and are compiled in the “AFM” subfolder.

p.25


Figure 7. Jensen cation plots. Major compositional fields (komatiitic, tholeiitic, and calc-alkalic

delineated by thick boundaries. Volcanic rock names delineated by gray dashed lines. After Rollinson (1993).

mg# Compositions – A general indicator of the degree of differentiation of an ultramafic or mafic rock is given by its mg# = MgO/(MgO+FeO(t)) (mole%). The average and range of mg# and the number of analyses on which these are based are listed in this column for each intrusion. If a particular intrusion has multiple cores which have distinct mg#’s, the average and range of each core is given.

Although most of the rocks studied here show varying degrees of deuteric alteration and metamorphic recrystallization, these processes should have little affect on the original (igneous) mg# of these rocks since Fe and Mg are not very mobile elements during alteration and metamorphism. However, the extent to which recrystallization has obscured original igneous textures can be problematic since the significance of whole rock mg# in an intrusive mafic or ultramafic rock is dependent on whether and to what degree the rock is a cumulate. Because mafic mineral phases (olivine or pyroxene) will have mg#’s 10-20% greater than the magmas from which they crystallized, two rocks crystallized from the same magma can have significantly different mg#’s depending on the extent to which they crystallized by cumulus processes. Generally, cumulate rocks are recognized by having an intergranular texture and commonly an

p.26


igneous foliation, typically defined by plagioclase. Of course, for strongly metamorphosed rocks, the original igneous texture is commonly obscured, thus making recognition of cumulus texture difficult or impossible. Nevertheless, average and range of mg# should serve to as a general indication of how evolved the various intrusion are. Multi-element Spider Diagrams – A common method for characterizing and comparing multiple compositional components in an igneous rock is to normalize various elemental concentrations to a standard composition. Most common standard compositions are estimates of primordial mantle, chondrite meteorite compositions, and MORB basalt. Normalization to the first two compositions give a general indication of how evolved samples are from primitive mantle compositions and normalization to chondrite compositions has the further advantage that it is based on actual measured compositions rather than estimated mantle compositions (Thompson, 1982). Generally, the elements used in spider diagrams include a variety of trace and minor elements most commonly reported in INAA, ICP-MS and XRF analyses. They are typically listed from most incompatible on the left to most compatible on the right. Recommendations for the specific elements used in spider diagrams and their order vary significantly. For this study, we have chosen to generally follow the element suite and chondrite normalizing composition recommended by Wood and others (1979) and listed in Table 5. Table 5. Chondrite normalizing values used to construct spider diagrams listed in their plotting

order (mostly from Wood and others, 1979; Yb from Thompson, 1982) Rb K Th Ta Nb Ba La Ce Sr Hf Zr P 1.88 850 0.04 0.022 0.56 3.6 0.328 0.865 10.5 0.19 9.0 500 Ti Sm Y Yb Lu Sc V Mn Cr Co Ni 610 0.203 2.0 0.22 0.034 5.21 49.0 1720 2300 470 9500

The descriptions reported in the intrusion attribute table first take note of the clustering and

parallelism of the various curves as this give some indication of the compositional homogeneity and consanguinity of the intrusion. On this latter point, rocks formed from a common, though perhaps variably fractionated, parent magma should show a general parallelism in their curves. This is because the specific zigzag pattern of the curve is generally inherited from the source of the parent magma, whereas its subsequent fractional crystallization results in only shifting the overall vertical position and increasing the slope of the curve.

The slope of the curve will become steeped with fractional crystallization due to the enrichment of the most incompatible elements (IE) and depletion of the compatible elements (CE). Typically the curve pivots on unity at Mn. The amount of enrichment of incompatible elements is taken as the slope and position of the curve from V to Ba and is reported as the average enrichment factor above chondrite for this segment of the curve. The elements to the left of Ba tend to have comparable (though often variable) degrees of incompatibility which results in

p.27


the curve commonly flattening out. If a parent magma was derived from normal chondritic mantle, IE enrichment should ideally show a smooth increase from right to left with a leveling out to the left of Ba. Any deviations from this smooth increase would be considered an anomaly. Spikes above the general smooth trend of the curve are positive (+) anomalies and those dipping below the curve are negative (–) anomalies. The causes of the anomalies are varied (e.g., nonchondritic source, varied depths or degrees of melting, varied depths (and mineral assemblages) driving fractional differentiation, mantle metasomatism, depletion of mobile elements during metamorphism, etc.) and are beyond the scope of this study to interpret. On the compatible element side of the curve, the amount of CE depletion is generally taken as the average depletion factor of Ni. Spider diagram plots for the various intrusions are compiled in the subfolder labeled Spider Diagrams within the GEOCHEMICAL PLOTS FOLDER. REE Diagrams – Chondrite normalization diagrams of only rare earth elements (REE diagrams) provide similar information as multi-element spider diagrams in that they give a general indication of the degree of fractionation from primitive compositions. The chondrite normalizing compositions used for these plots are listed in Table 6. Moreover, because of their more systematic incompatibility, they tend to display smoother curves. The clustering and parallelism of the lines give an indication of the compositional homogeneity and cosanguinity of the intrusion. Table 6. Chondrite normalizing values used in REE diagrams (from Taylor and McLennan,

1985).

La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu 0.367 0.957 0.137 0.711 0.231 0.087 0.306 0.058 0.381 0.0851 0.249 0.0356 0.248 0.0381

The patterns or shapes of the REE curves can vary because of minor differences in the

incompatibility of specific rare earth elements or groups of REEs with different mafic minerals (see Rollinson, 1993, Fig. 4.8). One element that can show dramatic variations relative to the other REE is europium (Eu). Because of its near compatible in plagioclase, Eu will commonly show a pronounced positive anomaly in rocks with cumulus plagioclase, whereas noncumulate rocks that have experience plagioclase fractionation may show a negative anomaly. In the column here, the degree of enrichment (= negative slopes) of three parts of the curve segments are described to give an indication of the overall curve pattern. The three groupings are: LREE: La -Nd, IREE: Sm-Dy, and HREE: Dy-Lu. . Positive sloped segments are taken to indicate depletion. The overall range of REE enrichment is reported as the average enrichment of end-member elements – La and Lu. Plots of REE diagrams for all the intrusions are compiled in the REE subfolder of the GEOCHEMICAL PLOTS FOLDER.

p.28


Tectonic Discrimination Diagrams - Several types of diagrams have been developed that attempt to discriminate the tectonic setting of mafic rocks by their geochemical attributes (Rollinson, 1993, chapter 5). We have chosen to plot our data on the four tectonic discrimination diagrams shown in Figure 8. These particular diagrams were chosen because most analyses include these elements and because these elements are relatively immobile during alteration and metamorphism. Intrusion data plotted on these diagrams are compiled in the ZrTiY, ZrTiSr, ZrNbY, and MnTiP subfolders in the GEOCHEMICAL PLOTS FOLDER.

All these diagrams have limits to their accuracy and applicability, as pointed out by Rollinson (1993). Most diagrams are generally inappropriate for use in describing cumulate rocks, especially those rocks containing cumulus mafic phases, oxide, or apatite. The possible exception is the Zr-Nb-Y plot since all these elements are incompatible with potential cumulus phases. The presence of cumulus phases commonly can be recognized by the data plotting outside the normal fields (OR in the attribute table). Also, some data points may plot outside the normal range because they include rather imprecise analyses for Zr, Y or particularly Nb.

The Zr-Ti-Y and Zr-Ti-Sr diagrams of Pearce and Cann (1973) are mostly applicable to primitive to moderately evolved subalkaline (tholeiitic) basalts with CaO+MgO in the range of 12-20%. Both plots are helpful in discriminating between island arc tholeiites, island arc calc-alkaline basalts, and mid-ocean ridge basalts (MORB). The Zr-Ti-Y diagram also distinguishes within plate basalts (ocean island or continental flood basalts). The Zr-Nb-Y diagram of Meschede (1986) is particularly useful in distinguishing enriched MORB associated with mantle plumes from normal MORB with its depleted signatures. Plume-related alkali basalts and tholeiitic basalts can also be distinguished. Unfortunately, many analyses do not include Nb and many of the reported values are quite imprecise. Lastly, the Mn-Ti-P diagram of Mullen (1983) is appropriate for basalt and basaltic andesite with SiO2 in the range of 45-54 wt%. This plot is particularly inappropriate for plotting cumulate rocks since mafic phases incorporate Mn, oxide takes up Ti, and P is a major component of apatite.

p.29


Figure 8. Tectonic discrimination diagrams for mafic rocks. Sulfide Mineralization

Macroscopic occurrences of sulfide mineralization observed in drill core or outcrop are described in this column. Visual estimates of modal abundance are qualified with the terms rare (<0.1%), trace (<0.5%), minor (0.5-2%), moderate (2-5%), and substantial (>5%). Also reported, where known, are the distribution of sulfide concentrations stratigraphically through the drill core, the textural occurrences of the sulfides (e.g., disseminate, irregular, veined, clotted, etc.), and the basic sulfide mineralogy listed in order of decreasing abundance.

Sulfide Saturation

The amount of sulfide in the rock is best quantified by analyses of weight percent of sulfur. The whole rock concentration of sulfur can be an important attribute in evaluating the extent of sulfide saturation of the mafic or ultramafic magma from which the rock crystallized, assuming

p.30


the rock has not experienced any post-crystallization addition or subtraction of sulfide. Base-metal sulfide and PGE mobility is problematic in rocks that have been strongly deuterically altered or metamorphosed since sulfide can be very soluble in high-T magmatic hydrothermal fluids (Hanley, 2005). However, experimental and empirical evidence indicate that platinum group minerals are relatively immobile under low grade metamorphic or hydrothermal conditions, even in the presence of Cl-rich fluids that are nevertheless capable of mobilizing S, Cu, Ni, and Au, can be relatively insoluble in the presence of the same fluids (Hanley, 2005).

Sulfur solubility in a magma positively correlates with the Fe content of the magma and negatively correlates with the temperature (Haughton and others, 1974). Figure 9 shows an idealized variation in S wt.% in a magma at sulfide saturation over a differentiated range of compositions from ultramafic to felsic. This curve is a schematic summary of sulfide saturation estimates based on experimental and natural systems (Naldrett and von Gruenewaldt, 1989; Li and others, 2001; Li and Ripley, 2005). The rapid drop in sulfur solubility during differentiation of ultramafic magmas is caused by rapidly decreasing Fe content due to crystallization of mafic phases and decreasing temperature. During fractionation of mafic magmas, sulfur solubility is roughly constant due to the offsetting effects of gradually increasing iron (Fenner trend) and decreasing temperature. At the intermediate to felsic end of differentiation, sulfur solubility decreases as Fe content declines along with temperature. Based on the general ranges of S shown in Figure 9, the S concentrations analyzed for the various intrusions listed in the attribute table are classified as indicating different degrees of sulfide saturation. These sulfide saturation groupings are listed in Table 7. The number of samples classifying into the different groupings are listed in the S wt% column of the intrusion attribute table.

Figure 9. Schematic diagram showing the idealized variation in sulfur concentration in

progressively differentiated magmas from ultramafic to felsic compositions. Based on data from Li and others (2001).

p.31


Table 7. Degrees of sulfide saturation based on concentration of S wt%. Sulfide Sulfide Sulfide Strongly Sulfide Undersaturated Saturated Oversaturated Oversaturated

Mafic <0.08 0.08-0.15 0.15-1.0 >1.0 Ultramafic <0.08 0.08-0.25 0.25-1.5 >1.5

Intrusions that are strongly altered or metamorphosed may have concentrations of sulfide that have little relationship to their initial magmatic concentrations and thus making the sulfide saturation classifications irrelevant. Since most primary sulfide over saturation occurs in the contact zones of intrusions due to footwall contamination, a clue that sulfide may be metamorphic or metasomatic is the occurrence of samples classifying as sulfide oversaturated that are far removed from contact zones or sulfide-bearing inclusions. Further evidence that sulfide is likely metamorphic is that it is dominantly Fe-sulfide. Sulfide Type

Primary sulfide associated with PGE mineralization is invariably dominated by Cu- and Ni-rich sulfide minerals of chalcopyrite, bornite, and pentlandite (Naldrett, 1989). In contrast, secondary metasomatic or metamorphic sulfide or PGE-depleted magmatic sulfide will be composed of Fe-sulfide (pyrite or pyrrhotite). Although general abundances of chalcopyrite were noted during core logging, another way to ascertain the amount of Cu-rich sulfide in a sample is to evaluate the S/Cu ratio of the whole rock. A rock containing only chalcopyrite (CuFeS2) should have a S/Cu atomic ratio of approximately two. Therefore, to get a general sense of the amount of Cu-rich vs. Fe-rich sulfide in the intrusions, the S(*1000)/Cu ratios were calculated for all samples and the sulfide mineralogy was categorized as being Cu-dominant (S/Cu <1.5), mixed Cu/Fe-sulfide (S/Cu 1.5-3.0), or Fe-dominant (S/Cu > 3.0). The number of samples in each sulfide type category for each intrusion are listed in this column of the intrusion attribute table. PGE+Au Grade

The maximum concentration of Pt+Pd+Au values measured in the various intrusions are listed in this column, with averages indicated for some cases intrusions with many analyses. Except for the contact zone mineralization of the Winterfire intrusion, which contains up to 1400 ppb combined Pt+Pd+Au, no other samples analyzed for this study exceed 400 ppb, and these values come almost exclusively from elevated Au concentrations. Indeed, the greatest Pd or Pt values outside the Winterfire intrusions do not exceed 40 ppb.

These low values are not necessarily discouraging from the perspective of assessing potential for economic PGE mineralization. Most mafic rocks that are crystallized from magmas that have not yet become sulfide saturated and consequently have not been scavenged of their precious

p.32


metals have whole rock PGE concentrations of less than 10 ppb combined PGE. The actual concentration will depend on the amount of PGE-bearing magma to PGE-barren cumulus minerals in each whole rock analysis (Barnes and others, 1997). Total PGE is in and of itself a poor indication of the potential for PGE mineralization. A better indicator is to evaluate the Cu/Pd ratio of the rock. PGE Depletion (Cu/Pd ratio)

As first pointed out by Barnes and others (1993), variations in the Cu/Pd ratio within a mafic intrusion can be a very useful indicator of the history of sulfide saturation and the potential for PGE mineralization. Because Pd has a silicate melt/sulfide liquid partition coefficient that is a one to two orders of magnitude greater than Cu (DCu(sul/sil) ~1000-1400, DPd(sul/sil) 20,000-35,000; Crockett, 2002; Mungall, 2005), a sulfide saturation event should result in a significant increase in the Cu/Pd ratio of the silicate magma. Barnes and others (1993) devised a Cu/Pd vs. Pd plot (Fig. 10) to illustrate the effect of sulfide saturation on the Cu/Pd ratio and PGE grade relative to what is estimated to be an undepleted mantle ratio (2x103 to 1x104) and a typical Pd concentration of 3-10 ppb of tholeiitic basalt. Upon sulfide saturation, the Cu/Pd ratio and Pd concentration in the crystallized rock will depend on the amount of sulfide accumulated with silicate and oxide minerals and the R-factor of the sulfide. The R-factor is the mass ratio of silicate melt to sulfide liquid. In other words, it is the amount of silicate melt that each aliquot of sulfide liquid “sees” and can potential scavenge of its PGE. If sulfide saturation occurs in a very dynamic manner, such as through magma mixing, this will promote a very high R-factor. If, on the other hand, sulfide saturation is passively triggered by fractional differentiation and boundary layer crystallization, this is likely to result in a low R-factor. Compositional curves leading out from the mantle value to the enriched field are shown for three different R-factors (102, 104, 105) and are scaled to different sulfide concentrations in Figure 10. Classic PGE reefs such as the Merensky Reef of the Bushveld Complex have Cu/Pd ratios and Pd concentrations that imply R-factors between 104 and 107 with sulfide abundances between 0.5 and 10 wt.%. Figure 10 also shows that residual magma that has experienced sulfide saturation will be driven to the upper left hand corner of the depleted field. Rocks crystallized from these residual magmas will have saturated sulfide concentrations, but depleted (high) Cu/Pd ratios and low PGE concentrations.

The Cu/Pd ratios of the various intrusions are inventoried in this column of the intrusion attribute table and classified as to whether they have enriched ratios (Cu/Pd < 2x103), mantle (undepleted) ratios (Cu/Pd = 2x103 - 1x104), slightly depleted ratios (Cu/Pd = 1x104 - 2x104), depleted ratios (Cu/Pd > 2x104). For many analyses, the measured concentrations of Pd are below detection limits, which for the various sources of data range from 0.1 to 2 ppb. Analyses that indicate Pd below detection limits are given the detection limit value or lower values, if other similar and more precise analyses exist. On the other hand, Cu analyses of all vintages tend to be relatively precise and well above detection limits of 1-5 ppm. Therefore, many of the stated Cu/Pd ratios may be underreporting the actual ratios such that some of the ratios are more depleted than indicated. The Cu and Pd data listed in GEOCHEMICAL_DATA.xls/ worksheet

p.33


GDT for various intrusions are plotted on Cu/Pd vs. Pd diagrams and compiled in the Cu/Pd-Pd subfolder of the GEOCHEMICAL PLOTS FOLDER.

Figure 10. Variation in Cu/Pd ratios and Pd concentrations (in ppb) in mafic rocks that

experience sulfide saturation. The red diamond is a possible composition of an undepleted mantle-derived tholeiite. Upon becoming sulfide-saturated, the rocks formed will developed enriched concentrations of Pd and decreased Cu/Pd ratios depending on the R-factor (silicate/sulfide melt ratio) and concentration of sulfide in the rock. Compositional trends for three different R-factors are shown. The PGE reef deposits field show the compositions of high grade PGE reefs, which imply R-factors greater than 104 and sulfide concentrations between 0.5 and 10 wt.%. The dashed line shows the depleted compositions of Cu/Pd and Pd of residual magmas produced from a sulfide saturation event.

Exploration Recommendations

The ultimate purpose of this study is to attempt evaluation of the PGE potential of mafic and ultramafic intrusions outside the Duluth Complex, using existing data and sample materials, supplemented with new analyses and analytical techniques. The mineralization parameters of sulfide saturation, dominant sulfide mineralogy, and PGE depletion, combined with what is known about the igneous stratigraphy of each intrusion, does allow us to make predictions with varying degrees of certainty as to whether and where within particular intrusions significant PGE

p.34


mineralization might be found. This column briefly summarizes those predictions for each intrusion.

A useful diagram to graphically summarize the various mineralization parameter listed in the intrusion attribute table is presented in Figure 11. This graph compares the Cu/Pd ratio against total S (wt.%) of data distinguished by dominant sulfide type. Nine compositional fields are delineated on the diagram and are defined by zones of enriched, mantle, and depleted Cu/Pd ratios and zones of sulfide undersaturated, sulfide saturated and sulfide oversaturated zones based on sulfur concentration. Figure 11A shows the sulfide saturation zones for mafic rocks whereas Figure 11B shows the expanded zone of sulfide saturation for ultramafic rocks. Figure 11A also shows data from the Winterfire intrusions, which displays the best PGE mineralization of any of the intrusions investigated in this study. Most of the data comes from the well mineralized contact zones of the intrusion and thus the data clusters in fields of enriched to mantle Cu/Pd ratios and sulfide over saturation. Cu-sulfide is the most abundant type in the mineralized zones. The mineralization parameter data for the various intrusions listed in GEOCHEMICAL_DATA.xls/ worksheet GDT are plotted on similar diagrams and compiled in the Cu/Pd-S subfolder of the GEOCHEMICAL PLOTS FOLDER.

p.35


Figure 11. Plot of Cu/Pd ratio vs. S (wt.%). Color fields distinguish enriched, mantle and

depleted Cu/Pd ratios and grayscale fields distinguish sulfide undersaturated, sulfide saturated, and sulfide oversaturated fields based on sulfur concentration. Plot A is appropriate for mafic intrusive rocks and show data from the mineralized margins of the Winterfire intrusion. Color symbols distinguish dominant sulfide types based on S/Cu ratios. Plot B shows the wider range of S concentrations indicating sulfide saturation for ultramafic rocks.

PPGGEE RReeeeff

Orthomagmatic Sulfide Saturation

E

Figure 12. Schematic of two main types of PGE deposits occurring in mafic intrusions.

Exxtteerrnnaall SSuullffuurrSSoouurrcc

CCoonnttaacctt--ssttyyllee CCuu--NNii--PPGGEE MMiinnee

ee

rraalliizzaattiioonn

p.36


In making these recommendations, it is necessary to understand the types of PGE ore deposits

being sought and how they might be recognized. Most of the world’s PGE deposits associated with mafic-ultramafic intrusions fall in to two basic categories: 1) contact-style Cu-Ni-PGE mineralization and 2) stratiform PGE “reef” mineralization. Other mineralization styles such as intrusive breccia mineralization, as exhibited by Lac des Iles, and marginal and footwall Cu-Ni-PGE mineralization of the Sudbury Complex, are not considered major deposit types because they appear to represent unique circumstances of mineralization. These two deposit types are schematically summarized in Figure 12 and will be briefly described below. Contact-style Mineralization - This type of PGE deposit is perhaps the most common type of PGE mineralization, but it typically does not yield the highest grades of PGE. Some well known examples of this type of deposit are Norils’k, Jinchuan, Platreef, Voisey’s Bay and the Duluth Complex (Naldrett, 1997). Most would best be characterized as PGE-rich Cu-Ni sulfide deposits as PGE’s are almost always a bonus metal. As the name implies, these types of deposits are found in the contact zones of intrusions and are generally thought to form by extensive sulfide contamination of undepleted magmas by sulfidic country rock. Commonly, the contamination appears to originate from country rocks in the dynamic environment of feeder zones of the intrusions. Sulfide saturation and oversaturation triggered by this contamination can result in large concentrations of sulfide accumulating in the contact zones. The PGE tenor of this sulfide will depend on the fluid dynamics of how this sulfide liquid mixes with the silicate magma (the R-factor) and on the original PGE concentration of this magma. Additional enrichment can occur if this sulfide fractionates to produce a Cu-PGE rich differentiate. As demonstrated by the Norils’k deposits, the size, tonnage and grade of contact style mineralization does not necessarily correlate with the overall size of the intrusion. Rather, other factors such as the openness of the magma system, the geometry of the intrusion, the composition of the parent magma, and the reactability of sulfidic country rock can play a controlling role.

Short of drilling into the contact zones of intrusions, it is difficult to predict from samples taken from the interior of large intrusions whether the margins might be mineralized. Part of this difficulty commonly stems from inadequate information about the size of the magma chamber and whether it formed as an open system with multiple episodes of magma influx or as a single input of magma that crystallized as a closed system. Assuming the open or closed nature of an intrusion could be determined, Table 8 attempts identify the ideal mineralization parameters found in the interior of an intrusion that would promote contact-style mineralization. In a closed system, if sulfur contamination occurred to produce contact-style mineralization, then the complete intrusion should remain sulfide saturated during its crystallization. Later crystallized and possibly differentiated rocks in the interior of the intrusion should become progressively dominated by Fe-sulfide with depleted Cu/Pd ratios. In an open system, magma filling the interiors of the intrusions may have become insulated from any potential S contamination that might have been encountered by the initially emplaced magmas forming the marginal zones or it

p.37


may show local episodes of saturation due to minor contamination or to differentiation-driven fractionation. If these later magma injections are basically the same magma that formed the intrusion margins, which should commonly be the case, then the identification of undepleted Cu/Pd ratios in these interior rocks at least holds out the possibility that if S contamination did affect the early injections, the magma had high concentrations of precious metals. If the recharging magmas can be shown to be variable, then it becomes unknowable whether contact-style mineralization may have occurred. Ultimately, the best determination of whether contact-style mineralization has occurred is to find and sample the intrusion margins.

Table 8. Ideal mineralization parameters observed in the interior of an intrusion that might indicate contact-style or PGE reef mineralization.

Potential for: Sulfide SaturationDominant Sulfide Cu/Pd Ratio

Comment Contact-style mineralization

• closed system saturated

Fe-sulfide

depleted

need external sulfur source

Contact-style mineralization

• open system • common parent magma

throughout

undersaturated or locally saturated

either

undepleted


Contact-style mineralization

• open system • variable parent magma

undersaturated or saturated

either

undepleted to depleted


PGE reef

• stratigraphically higher undersaturated

Cu-sulfide

undepleted

less predictable in open system

PGE reef

• stratigraphically lower saturated

Fe-sulfide

depleted

less predictable in open system

PGE Reef Mineralization – Stratiform PGE reef deposits hosted by mafic layered intrusions are best known for their occurrences in ultramafic-mafic intrusive complexes such as Bushveld and Stillwater (Naldrett, 1989). Such deposits are most commonly found near the transition from ultramafic to mafic cumulates and occur as 1-3 meter thick intervals enriched in PGEs (1-20 ppm) and trace to moderate amounts of sulfide (0.5-5 wt %). Similar, but lower grade, PGE reefs have also been found in the upper differentiated parts of tholeiitic mafic layered intrusions and are termed Skaergaard-type reefs (Miller and Andersen, 2002). Although a variety of models have been proposed for the origin of PGE reefs, most reef occurrences in both ultramafic/mafic and tholeiitic mafic intrusion are thought to be primarily orthomagmatic in origin and related to the saturation, exsolution, and settling of sulfide melt from silicate magma (Naldrett, 1989).

Ideal conditions for the orthomagmatic formation of economic PGE reefs are: • The parent magma is initially sulfide-undersaturated.

p.38


• The parent magma has a high initial PGE concentration and/or experiences a considerable amount of fractional crystallization to build up metal concentrations prior to sulfide saturation.

• The initial segregation of sulfide melt is triggered by a process (e.g., crystallization differentiation, cumulus phase changes, magma recharge, country rock assimilation, magma venting) that promotes a high R factor (Fig. 10).

Evaluating whether an intrusion might host a PGE reef first requires establishing that the intrusion has a significant thickness (>1km), displays cumulate textures, and is well differentiated. With metamorphic recrystallization obscuring original igneous textures and the limited stratigraphic sampling of many intrusions, it is difficult to know if these criteria are met for many of the intrusions in this study. Beyond this, determining where a PGE reef might occur within a layered intrusion requires monitoring the chemostratigraphy of mineralization parameters as shown in Table 8. For closed magma systems, determining whether a PGE reef might occur higher or lower in the igneous stratigraphy relative to a particular drilled interval can be straightforwardly determined by monitoring the Cu/Pd ratio and sulfide saturation levels. In more open systems, this exercise becomes more difficult due to the resetting of mineralization parameters caused by frequent recharge. In this case, a more complete profiling of the igneous stratigraphy is required to seek out possible PGE reef horizons, with particular emphasis focused on the transition from ultramafic to mafic portions of intrusion.

A potential problem in discovering economic grade PGE deposits in either style of mineralization is the effects of alteration, metamorphism, and deformation. Dissolution and mobilization of base metal sulfide and PGM by Cl-rich high temperature, deuteric fluids is well documented, although evidence suggests that platinum group minerals are not as reactive at lower temperature conditions (Wood, 2002; Hanley, 2005). Remobilization of precious and base metal sulfide mineralization in contact zones and in PGE reefs can also occur during regional metamorphism, as demonstrated by the PGE-bearing intrusions of the Fennoscandian shield (Iljina and Lee, 2005). Another, though perhaps more tractable, problem is posed by deformation of intrusions, which can cause considerable dislocation of mineralized intervals. This may account for the difficulty in tracing the well mineralized, but thin Winterfire intrusive sill(s).

In summary, although many intrusions listed here hold out some promise of hosting significant PGE mineralization, it is clear that additional drilling and geochemical sampling is required to better determine the potential of any intrusion or group of intrusions. It should be reemphasized that the 38 intrusion listed in the intrusion attribute table are only those of the 180 intrusions inventoried in this study that had significant amounts of drill core or surface exposure to warrant more detailed study. Further exploration of many of these “unstudied” intrusions may also be warranted, especially those having similarities to the select intrusions that show some potential. Moreover, geophysical interpretations indicate that many more mafic and ultramafic intrusions likely lie buried beneath the glacial drift (Fig. 3), and many of these may warrant investigation.

p.39


Field and Geochemical Study of the Deer Lake and Newton Lake Complexes

As a component of this project, two layered mafic complexes were investigated in some detail by the Natural Resources Research Institute. They are the Deer Lake complex in northeastern Itasca County, and the Newton Lake Formation in the Vermilion district of St. Louis County (Fig. 1). A third layered sequence in northwestern Minnesota—the Mentor mafic intrusive complex—was not evaluated in detail, as it is poorly represented by only a few drill cores and the waste-rock of a mined gas storage cavern (Jirsa and Boerboom, 1993). The Deer Lake complex and Newton Lake Formation consist of steeply dipping, tholeiitic to komatiitic, mafic to ultramafic sills. The sills are variably differentiated and complexly interfingered with volcanic host rocks, some of which appear to be the extrusive equivalent of the sills. The two complexes lie in a similar stratigraphic position along the northern-most part of the Archean Wawa subprovince, though they are separated by a region of complex faulting nearly 100 km wide. Interestingly, the Mentor lies in a similar stratigraphic position several hundred kilometers to the southwest.

Both the Deer Lake and Newton were investigated in the 1970’s by explorationists in search of base-metal sulfides. Exploration was terminated by discovery that the rocks contain only minor amounts of sulfide minerals. Although sulfide mineral content is important for viable Cu-Ni deposits, the scarcity of sulfides does not preclude the presence of PGE mineralization. This archived work in the Deer Lake complex produced a fairly extensive database of geochemical data, which is incorporated with about 98 new analyses (DEERLAKE06.xls). Most sills have low PGE values; however, PGE was locally concentrated in the basal parts of several sills, possibly as the result of sulfide saturation events. Maximum PGE assays are 84.1 ppb Pt, and 52.2 ppb Pd. In addition, several anomalous Au values (>100 ppb), including one at 2,399 ppb, occur within sheared gabbro. The study also resulted in a new, detailed geologic map of the complex (Severson and Jirsa, 2005). Work in the Newton Lake Formation was reported in the products of Phase 1, released to open-file (Jirsa and others, 2003b), and no further effort was expended during Phase 2. Assay results show very low PGE-Au contents, despite the discovery of a major sulfide horizon within one of the sills. Maximum values in the Newton Lake rocks are 32.8 ppb Pt, 11.1 ppb Pd, and 16.0 ppb Au. Further PGE-related investigation of the Newton Lake Formation is probably not justified. Diabase Dikes

Several suites of mafic dikes are represented on the current map iteration (Figure 3 and ArcMap project), which depicts nearly 3000 individual dike segments. The dikes vary in composition from diabasic and gabbroic, to lamprophyric. Although some dikes are exposed and were core-drilled locally, their trace beyond these known areas is based largely on geophysical maps. Both normally and reversely polarized dikes occur, having magnetic high and low signatures, respectively. The magnetic signature is a relative one—for example, dikes that cross

p.40


magnetically high terranes may appear by contrast to be lows, and thus could be inferred as reversely polarized, and yet they may have either polarity. Only those dikes that pass through subdued magnetic terranes and into high or highly patterned magnetic terranes can be determined with some certainty as having reversed vs. normal polarity. Furthermore, narrow linear lows crossing magnetically high terranes are not necessarily dikes. An example is in the moderately well exposed granitoid rocks of the eastern Quetico subprovince, where fracture zones cut and apparently oxidized magnetic wall-rock. In similar magnetic terranes where bedrock exposure is limited, this condition cannot be ascertained, and might locally have been misinterpreted in this depiction.

The absolute ages of individual diabasic dikes are poorly known. The large swarm of northwest-trending dikes in central and northern Minnesota is generally known as the Kenora-Kabetogama swarm of Paleoproterozoic age (2076±5, Buchan and others, 1996). The ages of dikes having other orientations are questionable. Dikes adjacent to the Duluth Complex and related rocks, and those that cut the complex are inferred to be Mesoproterozoic. Some dike trends were projected by geophysical maps into areas inferred to be underlain by rocks younger than the dikes. Dikes that trend northward are probably underrepresented on this depiction because of the potential for confusion with slight variations in anomaly signatures between adjacent N-trending flight lines. CONCLUSIONS

Generally, our assessment indicates that nearly all types of intrusions in this study have some PGE potential; however, ultramafic (komatiitic) and mafic (tholeiitic) intrusions that display some petrographic evidence of differentiation are the most promising. The alkalic rocks and intrusions having mixed magmatic compositions typically have less potential. Diabase dikes are considered to have little potential, as their magmatic emplacement into generally cool crust typically results in little interaction with wall-rocks and only minor differentiation. Intrusions that probably have little potential for platinum group elements are included in this report in an effort to flesh-out the portrayal of mafic magmatism in Minnesota. Many other intrusions are inadequately represented by drill core and outcrop to determine their potential, despite this effort—specifically targeted drilling is required to address their potential.

PGE potential has been assessed to varying levels of detail for 38 intrusions listed in the intrusion attribute table (INTRUSION_ATTRIBUTES.xls) and Figure 4, using a combination of petrologic and geochemical methods. This study explores the potential for two styles of PGE mineralization; reef-type and contact-type. The most telling indicators for the potential for reef-type PGE deposits are the presence of cumulate textures or other indicators of differentiation, sulfur (S) content, composition of the sulfide phase (Cu vs. Fe); and variations in the Cu/Pd ratio. Evaluating the significance of these mineralization parameters requires some control on the position within the igneous stratigraphy of the intrusion samples are acquired. Of the 38 select intrusions, the Winterfire, Warroad East, O'Brian Creek, Lac qui Parle 241, Wilmont, Cottonwood, Cobden, and BO-1 intrusions best display these attributes. Because this study relies

p.41


on existing drill cores for samples, and those cores intersected only a small portion of each intrusion, some potential may exist along strike or at depth in many intrusions, particularly where magmas may have incorporated S-rich country rocks. The Grygla intrusion, and perhaps the Strathcona, have cumulate textures and compositions permissive of this scenario. Contact-type mineralization is favored in intrusions that partially melt and incorporate elements of their wall rocks. It is particularly difficult to evaluate directly, as few drill holes in our data set intersect contact zones of large individual intrusions. Nevertheless, the UBD-3, Black River, Oaks West and South, Warroad North, Linden, and Tamarack intrusions have geochemical signatures implying some potential exists for contact-type mineralization. Geochemical data indicate that both types of mineralization are possible, but not clearly indicated, for the O'Brian Creek, Kanabec EC-10, P14-B, Meeker BKV, and Cottonwood intrusions.

Several factors mitigate against PGE potential, including the presence of metamorphic recrystallization; metasomatic, hydrothermal, and deuteric alteration; and strong deformation. None of these conditions prevents mineralization; rather, each has the potential to remobilize elements, and thus reduce the predictability of deposit grade, volume, and morphology. The Winterfire, ANA, O'Brian metagabbro, Mizpah, St. Louis 284, Oaks South, and many of the Warroad intrusions may be examples of these phenomena. The Winterfire intrusion contains the largest PGE contents of any analyzed for this study; however, mineably significant tonnages apparently have not been delineated—we suspect this is due in part to the effects of remobilization. REFERENCES CITED AND BIBLIOGRAPHY [Minor annotations by authors in brackets] Barnes,S-.J., Couture, J.F., Sawyer, E.W., and Bouchaib, C., 1993, Nickel-Copper occurrences in the

Belleterr-Angleirs belt of the Pontiac Subprovince and the use of Cu-Pd ratios in interpreting platinum-group element distributions: Economic Geology, v. 88, p. 1402-1418.

Bauer, R.L., 1981, The petrology and structural geology of the western Lake Vermilion area, northeastern Minnesota: Ph.D. thesis, University of Minnesota, 204 p.

Bauer, R.L., 1985, Geologic map of Norwegian Bay quadrangle, St. Louis County, Minnesota: Minnesota Geological Survey Miscellaneous Map Series, M-59, scale 1:24,000.

Beltrame, R.J., Chandler, V.W., and Gulbranson, B.L., 1982, Geophysical investigation of the Cedar Mountain Complex, Redwood County, Minnesota: Minnesota Geological Survey Report of Investigations 27, 20 p.

Boerboom, T.J., 1987, Tourmalinites, nelsonites, and related rocks (Early Proterozoic) near Philbrook, Todd County, Minnesota: MS thesis, University of Minnesota, Duluth, 212 p.

Boerboom, T.J., 1994, Alkalic plutons of northeastern Minnesota: in Southwick, D.L., editor, Short contributions to the geology of Minnesota: Minnesota Geological Survey Report of Investigations 43, p. 20-38. [includes Baudette, Bello Lake, Cook, Coon Lake, Gheen, Idington, Linden and satellites, Morcom, and Side Lake plutons]

Boerboom, T.J.(project mgr), 2001, Geologic atlas of Pine County, Minnesota: Minnesota Geological Survey Geologic Atlas Series C-13 [PCC and PRC series drill holes].

Boerboom, T.J., 1997, Characterization of the Franklin Peridotite: Minnesota Geological Survey contract report to Minnesota Department of Natural Resources, Division of Lands and Minerals, Hibbing, Minnesota. [or contact author]

p.42


Boerboom, T.J., Jirsa, M.A., Southwick, D.L., Meints, J.P., and Campbell, F.K., 1989, Scientific core drilling in parts of Koochiching, Itasca, and Beltrami Counties, north-central Minnesota, 1987-1989: Summary of lithological, geochemical, and geophysical results: Minnesota Geological Survey Information Circular 26, 159 p. [KIB and KDH-series’ drill cores]

Boerboom, T.J., Southwick, D.L., and Severson, M.J., 1999, Bedrock geology of the Aitkin 30 X 60 minute quadrangle, east-central Minnesota: Minnesota Geological Survey Miscellaneous Map M-99, scale 1:100,000.

Boerboom, T.J., Southwick, D.L., and Severson, M.J., 1999, Bedrock geology of the Mille Lacs Lake 30 X 60 minute quadrangle, east-central Minnesota: Minnesota Geological Survey Miscellaneous Map M-100, scale 1:100,000.

Brugmann, G.E., Naldrett, A.J., and Macdonald, A.J., 1989, Magma mixing and constitution zone refining in the Lac des Iles Complex, Ontario: Genesis of platinum-group element mineralization: Economic Geology, 84:1557-1573.

Buchan, K.L., Halls, H.C., and Mortenson, J.K., 1996, Paleomagnetism, U-Pb geochronology, and geochemistry of Marathon dykes, Superior Province, and comparison with the Fort Frances swarm: Canadian Journal of Earth Sciences, 33:1583-1595.

Cannon, W.F., Schulz, K.J., Daniels, D.L., Anderson, R., Chandler, V.W., Holm, D., Schneider, D., and Van Schmus, W.R., 2005, Geology of Precambrian basement rocks in Iowa and the southern parts of Wisconsin and Minnesota: in Easton, M., and Hollings, P. (eds.), Institute on Lake Superior Geology, 51st Annual Meeting, Nipigon, Ontario, v. 51, Part 1-Proceedings & Abstracts, p. 10-11.

Carr, M.J., 1999, Igpet99 for Win95/NT: Terra Softa Inc., Somerset, NJ. Chandler, V.W., Jirsa, M.A., and Morey, G.B., 1997, Mineral potential assessment of northern St. Louis

County, southeastern Koochiching County, and northeastern Itasca County, Minnesota: Minnesota Geological Survey Open-File Report 97-5, text 27 p., 9 plates at scale 1:62,500.

Crocket, J.H., 2002, Platinum-group element geochemistry of mafic and ultramafic rocks: in Cabri, L.J., ed., The geology, geochemistry, mineralogy, and mineral beneficiation of Platinum-group element: Canadian Institute of Mining, Metallurgy, and Petroleum, CIM Special Volume 54, p. 177-210.

Daggett, M.D., 3rd, 1980, The structure and petrology of the Cedar Mountain Complex, Redwood County, Minnesota: Unpublished M.S. thesis, University of Minnesota, 85 p.

Dahlberg, H., Peterson, D., and Frey, B., 1989, 1988-1989 Drill core repository sampling project: Archean Greenstone: Department of Natural Resources, Division of Minerals Report 266.

Day, W.C., 1982, Geology and geochemistry of the Vermilion Granitic Complex: Institute on Lake Superior Geology, v. 28, (International Falls, Minnesota), p 98-122. [contains analyses of Pelican Lake and Ash River intrusions and surrounding rocks]

Day, W.C., Klein, T.L., and Schulz, K.L., 1994, Bedrock geologic map of the Roseau 1X2 quadrangle, Minnesota, United States, and Ontario and Manitoba, Canada: USGS Miscellaneous Investigations Series Map I2358-A, scale 1:250,000.

Day, W.C., Southwick, D.L., Schulz, K.L., and Klein, T.L., 1990, Bedrock geologic map of the International Falls 1X2 degree sheet, Minnesota, USA and Ontario, Canada: USGS Miscellaneous Investigations Series Map I-1965B, scale 1:250,000.

Englebert, J.A., and Hauck, S.A., 1991, Bedrock geochemistry of Archean rocks in northern Minnesota: Natural Resources Research Institute Technical Report NRRI/TR-91/12, 200 p. [Refers to many geochemical studies and analyses in NW and NE Minnesota, including lamprophyres and other mafic-ultramafic rocks]

Ervin, C.P., and Mudrey, M.G., Jr., 1976, Extension of a northern Minnesota lamprophyre province by geophysical studies: Journal of Geophysical Research 81:4917-4922.

Frey, B.A., and Lawler, T.L., 1989, 1988-1989 Geodrilling Report: Minnesota Department of Natural Resources, Division of Minerals Report 264.

Frey, B.A., and Venzke, E.A., 1991, Archean drill core description and assay in Lake of the Woods County, Minnesota: Minnesota Department of Natural Resources, Division of Minerals Report 278.

p.43


Geldon, A.L., 1972, Petrology of the lamprophyre pluton near Dead River: in Sims, P.K., and Morey, G.B., eds., Geology of Minnesota: A centennial volume: Minnesota Geological Survey, p. 153-159.

Green, J.C., 1970, Lower Precambrian rocks of the Gabbro Lake quadrangle, northeastern Minnesota: Minnesota Geological Survey Special Publication Series SP-13, 96 p.

Green, J.C., Phinney, W.C., and Weiblen, P.W., 1966, Geologic map of Gabbro Lake quadrangle, Lake County, Minnesota: Minnesota Geological Survey Miscellaneous Map M-2, scale 1:31,680.

Green, J.C., and Schulz, K.J., 1982, Geologic map of the Ely quadrangle, St. Louis and Lake Counties, Minnesota: Minnesota Geological Survey Miscellaneous Map Series M-50, scale 1:24,000.

Griffin, W.L., 1967, Geology of the Babbitt-Embarrass area, St. Louis County, Minnesota: Unpublished Ph.D. dissertation, University of Minnesota, Minneapolis.

Hanley, J.J., 2005, The aqueous geochemistry of the Platinum group elements (PGE) in surficial, low-T hydrothermal, and high-T magmatic-hydrothermal environments: in Mungall, J.E., ed., Exploration for platinum-group element deposits: Mineralogical Association of Canada Short Course Series Volume 35, p. 35-56.

Haughton, D.R., Roeder, P.L., Skinner, B.J., 1974, Solubility of sulfur in mafic magmas: Economic Geology, v. 69, p. 451-467.

Hemstad, C.B., Southwick, D.L., and Ojakangas, R.W., 2002, Bedrock geologic map of Voyageurs National Park and vicinity, Minnesota: Minnesota Geological Survey Miscellaneous map M-125, scale 1:50,000.

Holm, D.K., Darrah, K.S., and Lux, D.R., 1998, Evidence for widespread ~1760 Ma metamorphism and rapid crustal stabilization of the Early Proterozoic (1870-1820 Ma) Penokean Orogen, Minnesota: American Journal of Science, v. 298, p. 60-81.

Holm D. K., Van Schmus, W. R., Mac Neil, L. C., Boerboom, T. J., Schweitzer, D., and Schneider, D., 2005, U-Pb zircon geochronology of Paleoproterozoic plutons from the northern mid-continent, U.S.A.: Evidence for subduction flip and continued convergence after geon 18 Penokean orogenesis: Geological Society of America Bulletin, v. 117, p. 259-275.

Iljina, M.J., and Lee, C.A., 2005, PGE deposits in the marginal series of layered intrusions: in Mungall, J.E., ed., Exploration for platinum-group element deposits: Mineralogical Association of Canada Short Course Series Volume 35, p. 75-96.

Jirsa, M.A., 1990, Bedrock geologic map of northeastern Itasca County, Minnesota: Minnesota Geological Survey, Miscellaneous Map M-68, scale 1:48 000.

Jirsa, M.A., and Boerboom, T.J., 1990, Bedrock geologic map of parts of Koochiching, Itasca, and Beltrami Counties, north-central Minnesota: Minnesota Geological Survey, Miscellaneous Map M-67, scale 1:250 000.

Jirsa, M.A., and Boerboom, T.J., 2003, Bedrock geology of the Vermilion 30’X60’ quadrangle, northeastern Minnesota: Minnesota Geological Survey, Miscellaneous Map M-141, scale 1:100 000.

Jirsa, M.A., and Boerboom, T.J., 1993, Scientific and exploration drilling in northwestern Minnesota: Minnesota Geological Survey Information Circular 40, 106 p. [NW series drill holes]

Jirsa, M.A., Boerboom, T.J., Chandler, V.W., and McSwiggen, P.L., 1991, Bedrock geologic map of the Cook-Side Lake area, St. Louis and Itasca Counties, Minnesota: Minnesota Geological Survey Miscellaneous Map M-75, scale 1:48,000.

Jirsa, M.A., and Chandler, V.W., 1997, Scientific test drilling and mapping in east-central Minnesota, 1994-1995: Summary of lithologic results: Minnesota Geological Survey Information Circular 42, 105 p. [EC-series drill holes]

Jirsa, M. A., Chandler, V. W., Lively, R.S., and Boerboom, T.J., 2003(a), Maps of bedrock geology and superimposed magnetic on gravity (SMOG) anomaly for east-central Minnesota: Minnesota Geological Survey Miscellaneous Map M-132, scale 1:200,000.

Jirsa, M.A., Chandler, V.W., and Runkel, A, 1999, Bedrock geologic map of northwestern Minnesota: Minnesota Geological Survey Miscellaneous Map M-92, scale 1: 200,000.

Jirsa, M.A., and Morey, G.B., eds., 2003, Contributions to the geology of the Virginia horn: Minnesota Geological Survey Report of Investigations 53, 135 p.

p.44


Jirsa, M.A., and Miller, J.D., Jr., 2004, Bedrock geology of the Ely and Basswood Lake (U.S. portion) 30’ x 60’ quadrangles, northeastern Minnesota: Minnesota Geological Survey Miscellaneous Map Series M-148, scale 1:100,000.

Jirsa, M.A., Miller, J.D., Jr., and Severson, M.J., 2006, Geochemical data from Phase 2 of the study geology, geochemistry, and PGE potential of mafic-ultramafic intrusions in Minnesota excluding the Duluth Complex: Minnesota Geological Survey Open-File Report OF-06-1 (digital release only).

Jirsa, M.A., Severson, M.J., Miller, J.D., Jr., Chandler, V.W., Boerboom, T.J., Lively, R.S., Keatts, M.J., and Holm, D., 2003(b), Geology, geochemistry, and PGE potential of mafic-ultramafic intrusions in Minnesota excluding the Duluth Complex: Phase 1: Contract report (digital only) Minnesota Geological Survey.

Jirsa, M.A., Southwick, D.L., and Boerboom, T.J., 1992, Structural evolution of Archean rocks in the western Wawa subprovince, Minnesota: refolding of precleavage nappes during D2 transpression: Canadian Journal of Earth Sciences 29:2146-2155.

Keatts, M. J., Jirsa, M.A., and Holm, D., 2003, Results of 40Ar/39Ar single-grain analyses of Precambrian mafic intrusions in northern and east-central Minnesota (abs): Institute on Lake Superior Geology Proceedings, 49th annual meeting, Iron Mountain Michigan, Volume 49, Part 1, p. 41-42. [includes dates from Tibbettes Brook (R-1), ANA, K-1, Linden, 264-10/2, B21, Black River (A1-1), UBD, and Oaks (B21-3) intrusions]

Klein, T.L., 1987, Geochemical data from the International Falls and Roseau Minnesota CUSMAP projects: U.S. Geological Survey Open-File Report 87-366.

Klein, T.L., 1988, Tabulated geochemistry of bedrock samples from exploration drilling in the Roseau and western part International Falls 1X2 degree quadrangles, northern Minnesota: U.S. Geological Survey Open-File Report 88-525, 91p.

Klein, T.L., 1994, Geochemical analyses of bedrock samples from drill holes on and near Red Lake Indian Reservation lands, northern Minnesota: U.S. Geological Survey Open-File Report 94-429.

Klein, T.L., Day, Warren C., Horton, R.J., and Clark, R.J., 1997, Mineral resource assessment of the US portion of the International Falls 1X2 quadrangle, northern Minnesota: USGS Bulletin 2044, 54 p and maps at 1:250,000 scale.

Klein, T.L., Day, Warren C., Horton, R.J., and Green, G.N., 1999, Regional bedrock mineral resource assessment of the Roseau 1X2 quadrangle, northern Minnesota: U.S. Geological Survey Miscellaneous Investigations Series Map I-2358-B, scale 1:250,000.

Le Maitre, R.W., Bateman, P., Dudek, A., Keller, J., Lameyre Le Bas, M.J., Sabine, P.A., Schmid, R., Sorenson, H., Streckeisen, A., Woolley, A.R., and Zanenttin, B., 1989, A classification of igneous rocks and glossary of terms. Blackwell, Oxford.

Li, C., and Ripley, E.M., 2005, Empirical equations to predict sulfur content of mafic magmas at sulfide saturation and applications to magmatic sulfide deposits: Mineralium Deposita, v. 40, p. 218-230.

Li, C., Maier, W. D., de Waal, S. A., 2001, The role of magma mixing in the genesis of PGE mineralization in the Bushveld Complex: thermodynamic calculations a new interpretations: Economic Geology, v. 96, p. 653-662.

Lively, R.L., Bauer, E.J., and Chandler, V.W., 2006, Maps of gridded bedrock elevation and depth to bedrock in Minnesota: Minnesota Geological Survey Open File Report OF-06-2, digital only maps, scale irrelevant.

MacTavish, A.D., 1999, The mafic-ultramafic intrusions of the Atikokan-Quetico area, northwestern Ontario: OGS Open File Report 5997, 154p.

Martin, D.P., Dahl, D.A., Cartwright, D.F., and Meyer, G.N., 1991, Regional survey of buried glacial drift, saprolite, and Precambrian bedrock in Lake of the Woods County, Minnesota: Department of Natural Resources, Division of Minerals Report 280.

Martin,D.P., Meyer,G.N., Cartwright, D., Lawler, T.L., Pastika, Jirsa, M.A., and Boerboom, T.J., 1989, Regional geochemical study of glacial drift drill samples over Archean granite-greenstone terrane in the Effie area, northern Minnesota; Department of Natural Resources Report 263, 323 p. [OB-series drill holes]

p.45


Martin, D.P., Meyer, G.N., Lawler, T.L., Chandler, V.W., and Malmquist, K.L., 1988, Regional survey of buried glacial drift geochemistry over Archean terrane in northern Minnesota: Department of Natural Resources, Division of Minerals Report 252, 387 p. [OB-series drill holes]

McCall, G.W., Nabelek, P.I., Bauer, R.L., and Glascock, M.D., 1990, Petrogenesis of Archean lamprophyres in the southern Vermilion Granitic Complex, northeastern Minnesota, with implications for the nature of their mantle source: Contributions to Mineralogy and Petrology, 104: 439-452. [Dead River pluton and Black Bay sill]

Meschede, M., 1986, A method of discriminating between different types of mid-ocean ridge basalts and continental tholeiites with the Nb-Zr-Y diagram: Chemical Geology, v. 56, p. 207-218.

Meints, J.P., Jirsa, M.A., Chandler, V.W., and Miller, J.D., 1993, Scientific core drilling in parts of Itasca, St. Louis, and Lake Counties, northeastern Minnesota: Minnesota Geological Survey Information Circular 37, 159 p. [CD, DD, CDC, and SLI-series drill cores]

Meyer, G.N., (project mgr), 1995, Geologic atlas of Stearns County, Minnesota: Minnesota Geological Survey Geologic Atlas Series C-10. [SC and SR series drill holes]

Miller, J.D. Jr., 2004, Petrology and PGE potential of the Greenwood Lake intrusion, central Duluth Complex, Lake County, Minnesota: Minnesota Geological Survey Report of Investigations 62, 94 p.

Miller, J.D., Jr., and Andersen, J.C.Ø, 2002, Attributes of Skaergaard-type PGE reefs: in Boudreau, A.E., ed., Extended Abstracts of the 9th International Pt Symposium, Billings, Montana, p. 305-308.

Miller, J.D. Jr., Green, J.C., Severson, M.J., Chandler, V.W., Hauck, S.A., Peterson, D.E., and Wahl, T.E., 2002, Geology and mineral potential of the Duluth Complex and related rocks of northeastern Minnesota. Minnesota Geological Survey Report of Investigations 58, 207p. w/ CD-ROM.

Mills, S.J., Southwick, D.L., and Meyer, G.N., 1987, Scientific core drilling in north-central Minnesota: Summary of 1986 lithologic and geochemical results: Minnesota Geological Survey Information Circular 24, 48 p. [1986-CUSMAP series]

Morey, G.B., McSwiggen, P.L., Kuhns, M.J.P., and Jirsa, M.A., 1985, Analytical results of the public geologic sample program, 1983-1985 biennium: Minnesota Geological Survey Information Circular 22, 56 p. [analyses of ML-1, SK-1]

Morey, G.B., and McDonald, L.M., 1987, Analytical results of the public geologic sample program, 1985-1987 biennium: Minnesota Geological Survey Information Circular 25, 59 p. [chem analyses of Side Lake pluton, HB-87, 60-22 metagabbro I175, CUS-17]

Morey, G.B., and McDonald, L.M., 1989, Analytical results of the public geologic sample program, 1987-1989 biennium: Minnesota Geological Survey Information Circular 29, 66 p. [analyses of Philbrook-PB, KIB-57, KDH-5]

Morey, G.B., and Meints, 2000, Geologic map of Minnesota: bedrock geology: Minnesota Geological Survey Statemap Series S-20, version 3, scale 1:1,000,000.

Morey, G.B., and Southwick, D.L., 1991, Stratigraphic and tectonic framework of the Early Proterozoic Penokean orogen in east-central Minnesota: in Ojakangas, R.W., ed., Precambrian geology of the southern Canadian Shield and the eastern Baltic Shield: Minnesota Geological Survey Information Circular 34, p. 42-48.

Mullen, E.D., 1983, Mno/TiO2/P2O5: a minor element discriminant for basaltic rocks of oceanic environments and it implications for petrogenesis: Earth and Planetary Science Letters, v. 62, p. 53-62.

Mungall, J.E., 2005, Magmatic geochemistry of the PGE: in Mungall, J.E., ed., Exploration for platinum-group element deposits: Mineralogical Association of Canada Short Course Series Volume 35, p. 1-34.

Naldrett, A.J., 1989, Stratiform PGE deposits in layered intrusions, in Whitney, J.A., and Naldrett, A.J., eds., Ore deposition associated with magmas. Reviews in Economic Geology v. 4, p. 135-165

Naldrett, A.J., 1997, Key factors in the genesis of Noril'sk, Sudbury, Jinchuan, Voisey's Bay and other world-class Ni-Cu-PGE deposits: implications for exploration. Australian Journal of Earth Sciences, v. 44, p. 283-315.

Naldrett, A.J. and von Gruenewaldt, G., 1989, Associations of platinum-group elements with chromite in layered intrusions and ophiolite complexes: Economic Geology v. 84, p. 180-187.

p.46


O’Brien, Hugh E., 1982, Petrology, geochemistry and magnetics of a lamprophyre dike intruded into the Saganaga tonalite batholith, northeastern Minnesota: Senior thesis, University of Minnesota, Minneapolis. (MGS Library)

Pearce, J.A., and Cann, J.R., 1973, Tectonic setting of basic volcanic rocks determined using trace element analyses: Earth and Planetary Science Letters, v. 19, p. 290-300.

Plumb, K.A., 1991. New Precambrian time scale: Episodes, vol. 14 (2), p. 139-140 Ripley, E.M., 1983, Sulfide mineralogy and sulfur isotope geochemistry of layered sills in the Deer Lake

Complex, Minnesota: Mineralium Deposita, v. 18, p. 3-15. Rollinson, H.R., 1993, Using geochemical data: evaluation, presentation, interpretation: Longman

Scientific & Technical, Essex, England, 352 p. Ruotsala, Albert P., and Tufford, Sarah P., 1965, Chemical analyses of igneous rocks: Minnesota

Geological Survey Information Circular 2, 87 p. [new and compiled analyses from High Lake hornblendite, SW Mn peridotites, mafic dikes in Saganaga and Snowbank plutons]

Schmitz, M.D., 1994, Origin and petrogenesis of the Early Proterozoic Kenora-Kabetogama mafic dike swarm: seniors thesis, Macalester College, St Paul, MN,

Schmitz, M. D., Bowring, S. A., Southwick, D. L., Boerboom, T. J., and Wirth K. R., 2006, High-Precision U-Pb geochronology in the Minnesota River Valley subprovince and its bearing on the Neoarchean to Paleoproterozoic evolution of the southern Superior Province: Geological Society of America Bulletin, v. 118, p. 82-93.

Setterholm, D.R., Morey, G.B., Boerboom, T.J., and Lamons, R.C., 1989, Minnesota kaolin clay deposits: a subsurface study in selected areas of southwestern and east-central Minnesota: Minnesota Geological Survey Information Circular 27, 99 p.

Severson, M.J., Miller, J.D., Jr., Peterson, D.M., Green, J.C., and Hauck, S.A., 2002, Mineral potential of the Duluth Complex and related intrusions: in Miller, J.D., Jr., Green, J.C., Severson, M.J., Chandler, V.W., Hauck, S.A., Peterson, D.M., and Wahl, T.E., 2002, Geology and mineral potential of the Duluth Complex and related rocks of northeastern Minnesota: Minnesota Geological Survey Report of Investigations 58, p. 164-200.

Severson, M.J., and Jirsa, M.A., 2005, Bedrock geology of the Deer Lake Complex and vicinity, Itasca County, Minnesota: Minnesota Geological Survey Miscellaneous Map M-165, scale 1:24,000.

Sims, P.K., 1985, Bedrock geologic map of west-central Vermilion district, northern Minnesota: USGS I-1529, scale 1:48,000.

Sims, P.K., 1973, Geologic map of western part of Vermilion district, northeastern Minnesota: Minnesota Geological Survey Miscellaneous Map Series M-13, scale 1:48,000. [B+W, covers more area in GRB to S and contains much more detail than Sims, 1985, but lacks outcrop]

Sims, R.K., Austin, G.S., and Ikola, R.J., 1965, Interpretation of Lake Washington magnetic anomaly, Meeker County, Minnesota: Minnesota Geological Survey Report of Investigations 4, 13 p. [Drill core NB-81-1]

Sims, P.K., and Mudrey, M.G., Jr., 1972, Syentitic plutons and associated lamprophyres: in Sims, P.K., and Morey, G.B., eds., Geology of Minnesota: A centennial volume, Minnesota Geological Survey, p. 140-152.

Sims, P.K., and Mudrey, M.G., Jr., 1978, Geologic map of the Shagawa Lake quadrangle, St. Louis County, Minnesota: U.S. Geological Survey, GQ-1423, scale 1:24,000.

Sims, P.K., and Southwick, D.L., 1980, Geologic map of the Soudan quadrangle, St. Louis County, Minnesota: U.S. Geological Survey GQ-1540, scale 1:24,000.

Sims, P.K., and Southwick, D.L., 1985, Geologic map of Archean rocks, western Vermilion district, northern Minnesota: USGS Miscellaneous Investigations Series Map I-1527, scale 1:48,000.

Southwick, D.L., 1972, Vermilion granite-migmatite massif: in Sims, P.K., and Morey, G.B., eds., Geology of Minnesota: A centennial volume, Minnesota Geological Survey, p. 108-119.

Southwick, D.L., 1994, Assorted geochronologic studies of Precambrian terranes in Minnesota: A potpourri of timely information: in Southwick, D.L., ed., Short contributions to the geology of Minnesota,

p.47


1994: Minnesota Geological Survey Report of Investigations 43, p. 1-19. [Philbrook age 1854±4 Ar/Ar]

Southwick, D.L., 1996, Archean Superior Province: in Sims, P.K., and Carter, L.M.H., eds., Archean and Proterozoic geology of the Lake Superior region, U.S.A., 1993: U.S. Geological Survey Professional Paper 1556, p. 6-13.

Southwick, D.L., 2002, Geologic map of pre-Cretaceous bedrock in southwestern Minnesota: Minnesota Geological Survey Miscellaneous Map M-121, scale 1:250,000.

Southwick, D.L., Boerboom, T.J., and Jirsa, M.A., 1998, Geologic setting and descriptive geochemistry of Archean supracrustal and hypabyssal rocks, Soudan-Bigfork area, northern Minnesota: Implications for metallic mineral exploration: Minnesota Geological Survey Report of Investigations 51, 69 p.

Southwick, D.L., Boerboom, T.J., and McSwiggen, P.L., 2001, Geochemical characterization of Paleoproterozoic volcanic and hypabyssal igneous rocks, east-central Minnesota: Minnesota Geological Survey Report of Investigations 57, 33 p. [Woodland lamprophyre dike]

Southwick, D.L., and Chandler, V.W., 1987, Mica-bearing olivine pyroxenite of possible lamproite-kimberlite affinity in central Minnesota, Economic Geology 82:212-217. [P14-B]

Southwick, D.L., and Chandler, V.W., 1996, Block and shear-zone architecture of the Minnesota River Valley subprovince: implications for late Archean accretionary tectonics: Canadian Journal of Earth Science v. 33, p. 831-847.

Southwick, D.L., Meyer, G.N., and Mills, S.J., 1986, Scientific core drilling in central Minnesota: Summary of lithologic and geochemical results: Minnesota Geological Survey Information Circular 23, 186p. [MGS drilling series’ AB, CC, LV, P, and 1985]

Southwick, D.L., Morey, G.B., and McSwiggen, P.L., 1988, Geologic map of the Penokean orogen, central and eastern Minnesota, and accompanying text: Minnesota Geological Survey Report of Investigations 37, scale 1:250,000.

Southwick, D.L., and Ojakangas, R.W., 1979, Geology of Minnesota: International Falls Sheet, Bedrock geology: Minnesota Geological Survey, scale 1:250,000.

Southwick, D.L., Setterholm, D.R., and Boerboom, T.J., 1990, Scientific test drilling in west-central Minnesota: Summary of lithologic and stratigraphic results, 1987-1988, and some preliminary geological conclusions: Minnesota Geological Survey Information Circular 31, 98 p. [MGS drilling in the APL, BPL, HS, HB, PBI-series]

Southwick, D.L., Setterholm, D.R., Runkel, A.J., Patterson, C.J., and Chandler, V.W., 1993, Scientific test drilling, 1989-1992: Descriptions and interpretations pertinent to the bedrock geology and Quaternary hydrogeology of southwestern Minnesota: Minnesota Geological Survey Information Circular 39, 63. [MGS PR, YB, and SWG series drill cores]

Streckeisen, A.L., 1976, Classification and nomenclature of igneous rocks. Earth Science Reviews 12, p.1-35.

Streckeisen, A.L., 1979, A chemical approximation to the modal QAPF classification of igneous rocks: Neues Yahrb. Mineral. Abh., v. 136, p. 169-206.

Taylor, S.R., and McLennan, S.M., 1985, The continental crust: its composition and evolution: Blackwell, Oxford.

Thompson, R.N., 1982, British Tertiary volcanic province: Scott. J. Geol., v. 18, p. 49-107. Thompson, R.N., 1984, Dispatches from the basalt front. 1. Experiments: Proc. Geol. Ass., v. 95, p. 249-

262. Van Schmus, W.R., 2006, University of Kansas, Lawrence, Personal communication regarding U-Pb zircon

age date of sample from drill hole BO-1 in Fillmore County, southeastern Minnesota. Vaillancourt, C., Sproule, R.A., MacDonald, C.A., and Hulbert, L.J., 2001, Potential for platinum group

elements mineralization in mafic-ultramafic intrusions in Ontario: #38, Project Unit 01-303, in Summary of Field Work and Other Activities 2001, Ontario Geological Survey Open-File Report 6070, p. 38-1 to 38-10. [COPY in MJ files]

p.48


Wood, D.A., Tarney, J. Varet, J., Saunders, A.D., Bougault, H., Joron, J.L., Treuil, M., and Cann, J.R., 1979, Geochemistry of basalts drilled in the North Atlantic by IPOD Leg 49: implications for mantle heterogeneity: Earth and Planetary Science Letters, v. 42, p. 77-97.

Yaghubpur, Abdolmajid, 1979, Preliminary geologic appraisal and economic aspects of the Precambrian basement of Iowa: thesis, University of Iowa. [Plates in MGS Library theses include maps (mag., bedrock topo) and cross-sections]

Zietz, Isidore, 1964, A magnetic anomaly of possible economic significance in southeastern Minnesota: US Geological Survey Circular 489, 5 p.

p.49

Outline for PGE Report

Documents

Transcript of Outline for PGE Report