[line]

TOC | Abstract | Introduction | Geologic Data | Database | Applications | Acknowledgements | References

Appendix A. Development and evolution of containment science activities...
Appendix B. Examples from complete mineralogic analysis...
Appendix C. Forms used for and examples of detailed petrographic analysis...
Appendix D. Examples of NTS database queries using SQL


Appendix A. Development and evolution of containment science activities for the Nevada Test Site at Los Alamos National Laboratory

The science of Containment integrates nuclear and conventional physics, explosion phenomenology, geologic knowledge, and past testing experience to ensure the safest conduct of nuclear tests (Brownlee, 1983; Brownlee, 1993). Separate Containment programs at LANL and at LLNL served to provide effective review of the prospectus for each proposed test, and the USGS served to provide specific review for geologic work within both programs, as well as to provide relevant research. Final approval of the Containment plan for each test was issued by the Containment Evaluation Panel, a committee of experts within each of the disciplines that constitute the science of Containment (Carothers, 1993).

At Los Alamos, Rick Warren began petrographic analyses in late 1980, initially to support the Yucca Mountain project, then in early 1981 to improve geologic characterization for containment of tests at Pahute Mesa. His work was inspired by the petrographic work of Frank Byers, initiated during the mid-1970's to support field mapping and Containment for the USGS. Frank's petrographic analyses, summarized in Byers et al., (1976a), demonstrated the powerful applicability of accurate, quantitative (modal) petrographic analysis to geologic problems within the SWNVF. Frank's analyses are based on extraordinarily large total counts (see Quinlivan and Byers, 1977 for examples), and coupled with his extensive experience and skill, provide exceptionally accurate analyses for major and minor constituents. Warren et al. (1984) demonstrated that accurate mineral chemistry and mineral identifications obtained by electron microprobe provided additional, equally powerful data to solve geologic problems. Warren has carried out a concerted campaign to characterize the mineralogy and mineral chemistry of all volcanic units within the SWNVF, particularly between 1980 and 1987. During the first year of this campaign, he developed techniques for the precise determination of minor and trace constituents (Warren et al., 1984; Warren et al., 1989b). Byers et al. (1976a) demonstrated that minor and trace constituents are very important petrographic data, but quantification of trace constituents had never before been attempted for routine optical analysis. The techniques for precise determination of minor and trace constituents require comprehensive documentation, and provide an extremely high level of quality assurance for petrographic analysis as an additional benefit.

The application of Warren's work resulted in the solution of several regional geologic problems concerning the SWNVF (Warren, 1983b). By the late 1980's, he was collaborating with David Sawyer and his colleagues of the Containment Research Group of the USGS, led by Paul P. Orkild, to address regional geologic problems within the volcanic field. With Sawyer and his colleagues contributing field work, age dates, and expertise in stratigraphy, an improved system of stratigraphy was developed for the SWNVF (Warren et al, 1989a), both formally for major units (Sawyer et al., 1994) and informally for all units (Ferguson et al., 1994).

During 1984, Sue Freeman began assembly of geological data into a non-relational database to improve its applicability to both Containment and regional geologic problems (Freeman et al., 1985). In 1987, Emily Kluk helped Rick to resume this effort, using commercial data storage and analysis software. Beginning in 1995, Greg Cole has transformed and reorganized the original electronic dataset into a widely accepted, standard form that allows widespread use of these geologic data for future Containment needs or for a variety of other possible applications.

 

Appendix B. Examples from complete mineralogic analysis of sample BH86N/33 and their application in detailed petrographic analysis

Anorthoclase was determined by point count from petrographic analysis of split BH86N/33(B. Within Appendix C are the form used to document methods of petrographic analysis as well as the completed form for split BH86N/33(B. Other splits identified in table loc_sam_split provide petrographic data from a glass covered thin section, an age date, and chemical analyses by both neutron activation analysis and by X-ray fluorescence. From table ma_gr_measure, four grains that contain anorthoclase (gr_comp_code equals AO) were intersected by the point count, and therefore represent this mineral within split B. The grain with a gr_ID of 27 consists entirely of anorthoclase, but grains 33 and 43 are intergrown with sanidine. Within Appendix C are the form used to document optical analysis for individual grains, as well as a single page of the completed form for split BH86N/33(B. Grain 57 has an area of .0178 mm2, and is intergrown with plagioclase that has an area of .0805 mm2. Whole counts are assigned to each particular grain for each mineral group of felsic, mafic, and Fe-Ti oxide minerals, as defined in database table comp_list. Counts for individual members of each group are apportioned fractional counts based on their relative areas, so anorthoclase 57 is assigned 0.18 counts and plagioclase 57 is assigned 0.82 counts. Using each gr_comp_ID, table ma_gr_comp_texture reveals that plagioclase 57 is wormy, or coarsely resorbed, whereas anorthoclase 57 has no unusual texture. Table ma_gr_measure reveals that energy dispersive and quantitative electron microprobe analyses are available for both anorthoclase and plagioclase 57, as described in the microprobe section.

With one exception, all intergrown grain components are described for each gr_ID represented in database table ma_gr_measure, regardless of the method specified for the mineral in table pa_measure. Only grain components >0.00045 mm2 are always represented where a grain count, method C in database table pa_method, is applied; grain components <0.00045 mm2 are represented only when they are the largest for a grain. No phases other than felsic minerals are assigned the same gr_ID as grains that contain anorthoclase, so no other phases are associated with anorthoclase in split BH86N/33(B. Usually anorthoclase is strongly associated with other feldspars, so anorthoclase has an unusual occurrence within sample BH86N/33.

Table ma_clast_measure describes a lithic with a thin section area of 2.6 mm2, assigned a clast_ID of LI4 in split BH86N/33(B, as a lava of the volcanics of Stonewall Mountain. Within Appendix C are the form used to describe clasts by optical petrography as well as the completed form for split BH86N/33(B. Database table ma_clast_alt describes the alteration of this clast as microgranophyric, a type of high-temperature devitrification, as defined in table alt_list. In table ma_clast_measure, a clast_pa_meth_code of B indicates that all grain components within LI4, except for Fe-Ti oxides, are represented in table ma_gr_measure. The minerals listed within this lithic include anorthoclase, apatite, biotite, monazite, plagioclase, quartz, and sanidine; many of these minerals, the table shows, have been analyzed by electron microprobe.

 

Appendix C. Forms used for detailed petrographic analysis, followed by examples for sample BH86N/33. Use of these forms requires a photomap, as shown in Figure 5. Optical analyses have been corrected for errors in original optical identifications discovered by microprobe analyses during December 1997. Areas are shown for examples as they are entered into forms, for convenience and clarity, in units of 0.0001 mm2.

Point Count, petrographic methods, and notes for split_ID:

Meth

Magnif

light

grid

TS area

 

Meth

Magnif

light

grid

TS area

                     
                     
                     

 

Min ID

Meth

Total pts

Min pts

Min ppm

Rank

 

Min ID

Meth

Total pts

Min pts

Min ppm

Rank

                         
                         
                         
                         
                         
                         
                         
                         
                         
                         
                         
                         
                         
                         
                         
                         
                         

ppmV = 0 method:                        components:

 

 

Lithology, alteration:

TS Narrative:

Clasts in split_ID:                                                    page__ of__

Clast ID

Clast code

Clast area

Area meth

Geol unit

Lith

Altn

Meth

Clst cts

Anlyst

Anal date

                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     

 

Minerals in split_ID:                                                   page__ of__

Grain ID

Clst PA

Host ID

Min ID

Min area

Textur

Min Host

Min cts

Pr ver

Anlyst

Anal date

                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     
                     



Point Count, petrographic methods, and notes for split_ID:  BH86N/33(B

Meth

Magnif

light

grid

TS area

 

Meth

Magnif

light

grid

TS area

1

500

RT

1.17576

343

           

 

Min ID

Meth

Total pts

Min pts

Min ppm

Rank

 

Min ID

Meth

Total pts

Min pts

Min ppm

Rank

SD

1

292

51.04

174800

1

AP

A

327

1

AO

1

292

1.975

6765

1

 

BT

A

   

2186

1

PL

1

292

6.972

23880

1

 

CX

A

   

826

1

VP

1

292

41

140411

1

 

OL

A

   

150

1

LI

1

292

2

6849

2

 

SN

A

   

0.3

1

VO

1

292

29

99315

1

 

PE

A

   

20

1

BT

1

292

1

3425

2

 

PO

A

   

0.3

1

CX

1

292

2

6849

2

 

ZR

A

   

129

1

MT

1

292

2

6849

2

 

IL

E

   

290

1

QZ

1

292

1

3425

1

 

MT

E

   

1575

1

LI

L

   

8413

1

 

HN

L

   

1000

1

ppmV = 0     method: 1        components: HN,OX,OL,NA,IL,AY,SU,MM
                                  A                           OX,NA,SN,AL,MN,PY,MM

Lithology, alteration: MG/VP

TS Narrative: Abundance for HN from grains examined is 824 ppmV.

 

Clasts in split_ID:  BH86N/33(B                                page 1 of 1

Clast ID

Clast code

Clast area

Area meth

Geol unit

Lith

Altn

Meth

Clst cts

Anlyst

Anal date

LI1

LI

1008

S

 

LA

GS

A

1

RGW

23Aug97

LI2

LI

688

S

 

VOL

QC

A

0

 

6Sep97

LI3

LI

836

S

     

A

0

 

24Aug97

LI4

LI

25760

S

Ts

LA

MG

B

1

 

13Dec97

LI5

LI

565

S

 

GR

 

A

0

 

7Sep97

 

 

Minerals in split_ID: BH86N/33(B                        page 7 of 10

Grain ID

Clst PA

Host ID

Min ID

Min area

Textur

Min Host

Min cts

Pr ver

Anlyst

Anal date

1

   

SD

762

   

1

Y

RGW

23Aug97

1

   

GZ

1

M

 

0

N

RGW

23Aug97

2

   

SD

390

   

1

Y

RGW

23Aug97

O1

 

PC1

MT

67

A

 

1

N

RGW

23Aug97

VP1A

   

VK

396

M

 

1

Y

RGW

23Aug97

27

 

PC

AO

1390

   

0.48

Y

RGW

14Dec97

27

 

PC

AO

1489

   

0.52

Y

RGW

14Dec97

33

 

PC2

AO

8560

   

1.39

Y

RGW

22Dec97

33

 

PC2

SD

3720

   

0.61

Y

RGW

22Dec97

43

   

AO

1010

 

SD

0.19

Y

RGW

22Dec97

43

   

SD

4350

   

0.81

Y

RGW

22Dec97

57

 

PC

PL

805

W

 

0.82

Y

RGW

22Dec97

57

 

PC

AO

178

   

0.18

Y

RGW

22Dec97

VP3

A

LI4

AO

309

R

 

1

Y

RGW

13Dec97

A4

A

LI4

BT

80

F

 

0

N

RGW

25Aug97

A4

A

LI4

AP

1

 

BT

0

N

RGW

25Aug97

A4

A

LI4

AP

1

 

BT

0

N

RGW

25Aug97

B4

A

LI4

BT

68

F

 

0

N

RGW

25Aug97

C4

A

LI4

MN

12

   

0

N

RGW

13Dec97

D4

A

LI4

QZ

235

   

0

Y

RGW

25Aug97

E4

A

LI4

QZ

565

   

0

Y

RGW

25Aug97

F4

A

LI4

AO

421

R

 

0

Y

RGW

13Dec97

I4

A

LI4

PL

590

R

 

0

Y

RGW

13Dec97

G4

A

LI4

PL

347

A

 

0

Y

RGW

25Aug97

G4

A

LI4

SD

186

A

 

0

Y

RGW

25Aug97

H4

A

LI4

PL

48

A

 

0

Y

RGW

13Dec97

J4

A

LI4

PL

26

A

 

0

Y

RGW

13Dec97

K4

A

LI4

SD

11

A

 

0

N

RGW

25Aug97

O7

   

MT

25

A

 

0

N

RGW

24Aug97

O7

   

ZR

1

   

0

N

RGW

24Aug97

NE/26

   

AP

111

   

0

Y

RGW

2Sep97

NE/26

   

AP

31

   

0

N

RGW

2Sep97

U

   

CX

167

 

HN

0

Y

RGW

13Dec97

U

   

MT

45

A

 

0

N

RGW

2Sep97

U

   

ZR

13

   

0

Y

RGW

2Sep97

U

   

HN

15

E/M

 

0

Y

RGW

13Dec97

AL

   

HN

34

M

 

0

Y

RGW

14Dec97

AO

   

HN

35

M

 

0

Y

RGW

14Dec97

 

Appendix D. Examples of NTS database queries using SQL (Structured Query Language)

1. Petrographic Data

Description of Analysis

Accessory minerals other than apatite and zircon are consistently present within certain stratigraphic units and are consistently absent within others. In many instances, the presence or absence of a key accessory mineral provides the most definitive difference between two strongly associated units. The most well known example is the presence of sphene in Ammonia Tanks Tuff (Tma) in contrast to its absence in Rainier Mesa Tuff (Tmr), as described by Byers et al. (1976a). Another striking difference between these two units is the presence of monazite in Rainier Mesa Tuff in contrast to its absence in Ammonia Tanks Tuff (Warren et al., 1989b). The monazite contents of these units, determined by detailed petrographic analyses, is shown in Figure D-1 using analyses extracted from the database.

Important considerations in this comparison are stratigraphic unit, sample type, concentration units, and petrographic method. Selecting stratigraphic units Tma and Tmr and all daughters (e.g., Tmrpl) ensures proper selection of the desired unit and all subunits, but includes mafic tephras at the base of Tmr that are inappropriate for comparison (Tmrd and Tmra), which are eliminated from the comparison set below. Volume concentrations are representative only within certain sample types, such as representative cuttings (sam_type_code DA in database table sample), core (code C), and Hunt sidewall (HS), and not within others such as mineral separates from core (code CM). Certain petrographic analyses yield semiquantitative results expressed as the number of grains observed (units code G in database table pa_measure), and so are not included in the comparison of quantitative results below. Finally, uncertainties in quantitative analyses for accessory mineral abundances depend strongly on the method of analysis; uncertainties are narrowest for pa_meth_code A in database table pa_measure. For most comparisons similar to that shown below, selection of data should consider the method of petrographic analysis, but all analyses for monazite currently within the database have been determined via pa_meth_code A.

Figure D-1. Characteristic monazite abundances applied to distinguish between Ammonia Tanks Tuff and Rainier Mesa Tuff (Tma) and Rainier Mesa Tuff (Tmr).

Description of Query

This query extracts petrographic abundance data for certain sample types of a given mineral within a particular stratigraphic unit. The query extracts data from three database tables: pa_measure, pa_split, and sample, which must be joined through common fields: spl_id and sam_id. A set of nested subqueries is used to identify petrographic splits for samples from the appropriate stratigraphic unit: Tma, as well as any subunits. These subunits are obtained through the parent-child relationship: strat_group_code:strat_code. The query and query results are provided in Figure D-2.

Figure D-2. Selection of petrographic measurements using nested subqueries and parent-child relationships (hierarchical queries).

 

2. Mineralogic Data

Description of Analysis

Mineral compositions determined by electron microprobe generally fall within narrow ranges for each stratigraphic unit (Warren et al., 1983b; Warren et al., 1989b). One important use of such mineral compositions is to define and correlate stratigraphic units, as described in section IV. The most useful mineral for such purposes is K-spar, which occurs within most volcanic units of the SWNVF. The K-spar of volcanic units of the SWNVF is invariably sanidine, but occasionally anorthoclase also occurs. Figure D-3 compares the combined potassium and barium contents of K-spar, expressed as mole% orthoclase (OR) plus celsian (CN) end member contents, between Tmr and Tma. Except for a single sample confidently assigned as Tmr (U19AB-600D), all samples of Tmr contain dominant sanidine within a narrow range from an OR+CN content of 62 mol%, as shown below. In comparison, all samples of Tma contain dominant sanidine with an OR+CN content centered at 47 mol%, and additionally anorthoclase with an OR+CN content centered at 22 mol%. Some K-spar compositions do not match dominant compositions owing to incorporation of xenocrysts from underlying units or to strong Na-enrichment (K-depletion) of rims by alkali diffusion during cooling, producing albitic rims as described by Warren et al. (1984). As seen in Figure D-3, albitic rims are typically observed for sanidine of Tmr, but not for Tma.

Stratigraphic units are exactly as for the query above. In contrast, sample type does not affect K-spar chemistry, and so need not be considered. The quality of the petrographic analysis is also irrelevant, so the sample set can include analyses from petrographic splits with acceptable analyses (table pa_split), or splits with unacceptable analyses (table xx_pa_split). Because Ba proxies for the dominant K, it is important to exclude analyses of K-spar lacking Ba; K-spar analyses require analysis of K (Table 9).

The acceptance criteria for microprobe analyses (Table 9) ensure accurate analyses. Analyses that do not meet these criteria can lead to erroneous conclusions. Mills (1991), using analyses from database table xx_probe_measure that violate these criteria, concluded that K-spars from Tma and Tmr are compositionally indistinguishable.

Figure D-3. Characteristic sanidine compositions (OR + CN) for the Ammonia Tanks Tuff (Tma) and Rainier Mesa Tuff (Tmr) which allow their mineralogic differentiation.

Description of Query

This query extracts measurements of combined mineral abundances for certain grain components within a particular stratigraphic unit. The query extracts data from table probe_end_members. A complex subquery containing two nested sets of subqueries is necessary to identify grain components for samples from the appropriate stratigraphic unit: Tma, as well as any subunits. The two sets of subqueries extract split ids (spl_id) from tables: pa_split and xx_pa_split (superceded or erroneous split information) to identify the full set of grain components which may have good mineral abundance measurements. These sets of split ids are combined through a "UNION" and then used to identify grain components which have measurements for either SD, AO or KF. Grain components identified by the subquery are then used to extract measurement values (end_member_value) from the table probe_end_members for end members: OR and CN. As there is a requirement that both OR and CN values be present in order for the grain component to be included in the analysis, both of these end member values are extracted from the database. These end member values are summed by GROUPing results by grain component and extracting only those grain components HAVING both end members present (COUNT = 2). The query and query results are provided in Figure D-4.

Figure D-4. Selection of summed mineralogical measurements using unions, nested subqueries and parent-child relationships (hierarchical queries).

 

3. Chemical Data

Description of Analysis

At Yucca Mountain, samples of mafic-poor Calico Hills Formation (Thp) that are zeolitic and have major amounts of clinoptilolite are often strongly depleted in Na and enriched in Ca compared to samples that are not zeolitic (Broxton et al., 1987). At Pahute Mesa, north across Timber Mountain from Yucca Mountain, similar samples of Thp are generally only slightly depleted in Na and enriched in Ca, as seen in Table D-1. For discussion of the correlation of Thp across Timber Mountain, see Figure 13 and discussion in text. This chemical difference measures the extent of cation exchange, Ca for Na, in clinoptilolite within the two regions.

Important considerations in this comparison are stratigraphic unit, sample type, and location of the sample. The selection of stratigraphic unit, Thp, is simple. Similar to the first query, only representative sample types provide chemical values appropriate for comparison. A simple geographic demarcation divides the analyses into those from Pahute Mesa, which have UTM northings >4100000 m, versus those from Yucca Mountain, which have northings <4100000 m.

Among other possible considerations are analysis type, a more detailed consideration of sample type, and treatment of analyses below detection limits. Almost all analyses for this comparison were obtained by X-ray fluorescence (XRF) or by neutron activation analysis (NAA). For the elements compared, uncertainties that accompany most of the analyses summarized below in database table ca_measure indicate that XRF analyses are generally superior. The samples within each geographic set include outcrop and subsurface, which might have a significant chemical effect. Although no analyses for the three elements below are below detection limits, many analyses for MgO fall below detection limits, so averages for MgO must consider such values. Because detection limits for NAA analyses of MgO are very poor, such analyses are probably best ignored, but XRF analyses are very sensitive for MgO, and values below detection limits certainly must be treated. Values below detection limits can be included to provide reasonably accurate averages when a relatively small fraction of the analyses (certainly <25%) falls below such limits; a value of half the detection limit is typically assumed for analyses below such limits. To evaluate differences for MgO between Pahute Mesa and Yucca Mountain clearly requires familiarity with chemical analyses. This example illustrates the following caveat: it is very important for the user of this database to be familiar with analytical methods for each type of data used, and with the types of samples valid for the planned application.

Table D-1. Average chemical values and their uncertainties for zeolitic mafic-poor Calico Hills Formation. Values from each site are averaged to provide a single value that is averaged with values from other sites. Uncertainties are twice the standard error of the mean.

Location

Sites

CaO

K2O

Na2O

   

avg

unc

avg

unc

avg

unc

Pahute Mesa

8

1.14

0.44

4.22

0.59

2.56

0.72

Yucca Mountain

55

1.79

0.14

4.83

0.36

1.12

0.13

 

Description of Query

This query extracts measurements of average chemical abundances for certain for geographic locations within a particular stratigraphic unit. The query extracts data from tables: ca_measure, and location, using table loc_sam_split to relate measurements to locations. A nested subquery identifies chemical splits from the appropriate stratigraphic unit: Thp. The ca_measure table contains NULL value entries for oxide_value measurements that failed to identify the presence of a chemical. For this case, a quantity equal to one half of the Lower Detection Limit (oxide_ldl) for the particular analysis and/or measurement technique is used to quantify the chemical abundance. In order to extract abundances for all measurements including NULLs, and then assign an average value at a geographic site, a set of temporary "views" is created. The first two views (temp1, temp2) extract non-NULL and NULL measurements, respectively. View temp3 combines the results of temp1 and temp2 for the final select statement in which the measurements are grouped by location and averaged. The query is provided in Figure D-5a and the results are provided in Figure D-5b.

Figure D-5a. Query for selection of averaged chemical measurements for geographic sites using views with unions and nested subqueries.

Figure D-5b. Results for selection of averaged chemical measurements for geographic sites using views with unions and nested subqueries.

 

Table 9. Acceptance criteria for microprobe analyses within this database. Criteria have been developed during 20 years by Rick Warren, based on published analyses (Deer et al., 1962a-c; Papike, 1987, 1988) and practical limits for microprobe analysis. Criteria have not been applied rigorously, but >99% of the acceptable analyses within this database conform. Acceptable analyses must include essential elements, with a few exceptions denoted by asterisks allowed within the database. Monitor elements are undetectable by microprobe analysis, but are usually associated with alteration or with secondary fluorescence of adjacent minerals; analyses for such elements above the indicated values indicates a problem and the analysis should be considered unacceptable. Values shown for monitor elements apply to oxide concentrations; asterisks denote elements that are poor monitors, but if analyzed, should provide the value shown. The molecular basis for structural criteria is 24 oxygens, except for hornblende, based on 23, and biotite, based on 22. Tetrahedral cations (t) are Si, Al, P, Nb, and Ti in glass only, and octahedral cations (o) are all other cations. The charge ratio R is the ratio of charge deficiency within the tetrahedral site to the total charge of cations within the octahedral site. The charge deficiency within the tetrahedral site is the sum of trivalent cations minus the sum of pentavalent cations. Concentrations of FeII and FeIII in magnetite and ilmenite are calculated from Spencer and Lindsley (1981); values are allowed to deviate from their structural basis, shown below, only when all Fe occurs as FeII. The structural criterion for glass is based on Warren (1983a).

 

elements analyzed as oxides

Acceptable

Structural

Mineral

essential

recommend

monitor

Totals (%)

Criteria

           

feldspar

Na,Al,Si,K,Ca,FeIII*,Ba*

Sr

Mn*,Ni*,Cr* < 0.1%

Ti* < 0.3%

97.5-102

o 2.85-3.15 (except adularia)

R 0.95-1.05

biotite

Na,Mg,Al,Si,K,Ti,FeII,Ba*

F,Cl,Mn

Ca < 0.2%

92-102.5

o 6.5-8.5, t 7.5-8.7

K2O+0.614 X BaO > 7%

hornblende

Na*,Mg,Al,Si,K,Ca,Ti,FeII

F,Cl,Mn

 

94-102

CaO 9.5-13.5%

clinopyroxene

Na,Mg,Al,Si,Ca,Ti,Mn*,FeII

Cr

K* < 0.1%, Ba* < 0.3%

97-102.5

o < 8.4

t > 7.9

orthopyroxene

Mg,Al,Si,Ca,Ti,Mn*,FeII

Na,Cr

K* < 0.1%, Ba* < 0.3%

97.5-103

o < 8.3

t > 7.85

olivine

Mg,Si,Ca,Mn*,FeII

Ni,Zn

Na*,Al,P*,K* < 0.1%

Ti < 0.2%, Ba* < 0.3%

98-104

o 11.6-12.4

t 5.8-6.2

ilmenite

Mg,Al,Ti,Mn*,FeII,FeIII

Cr,Zn,Nb

Si,Ca < 0.2%, Zr < 0.3%

96-103

o 6, t 12

magnetite

Mg,Al,Ti,Mn*,FeII,FeIII

V,Cr,Zn

Si,Ca < 0.2%, Zr < 0.3%

96-103

o 8, t 8

glass

Na,Mg*,Al,Si,K,Ca,Ti*,FeIII

F,P,Cl,Mn,Ba

 

<102

R >0.95

zeolite

Na,Mg*,Al,Si,K,Ca

FeIII,Ba

Mn* < 0.1%, Ti* < 0.2%

<102

R 0.85-1.05

silica

Na,Al,Si,K*,Ca,FeIII

 

Cr*,Mn*,Ni* < 0.1%, Ti < 0.2%

97-102.5 (except OP <99)

 


[back arrow]Back