view file

Canadian Mineralogist
Vol. 12, pp. 346-351 (1974)
SOLID SOLUTION IN THE WAGNERITE STRUCTURE*
KEUNHO AUH aNp F. A. HUMMEL
CeramicScienceSection,Material SciencesDepartment,The PennsylvaniaState University,
University Park, Pennsylvania16802
AssrRAcr
Complete substitution of Co++, Ni++, Cu+',
and. Zt++ for Mg++ was made in the wagnerite
structure. A complete solid solution series exists
between the four derivatives and Mg2(POJF, pro"
ducing linear changes in peak positions in the .tray diffractiotr patterns. Co, Ni, and Cu are also
completely soluble in *zinc wagnerite" and ZnsMg
(POJrF2n but in these cases non-linear changes in
peak positions are obseryed. Ca++ was not soluble in Ni or Cu wagnerite, but 15 and 30 mole Vo
Ca were soluble in MglPOa)F and CozGOJF, respectively. Exploratory runs of Li+, V5+, and Cl'
solubilities showed that 0.6 mole of Li could be
substituted in Mg, Co, and Ni wagnerites to give
Li0.ol113.aProsF1.aE6.s.
Approximately 4O mole Yo
V5+ could be substituted for P5+ in mapesium,
cobalt, and nickel wagnerites. Complete substitution of Cl' for F- was possible in Cu and Zn wagnerites, but not in Mg, Co, and Ni analogues. The
synthesis of the charge-coupled wagnerites LiMg3
PSOsF2 and NaMggPSOsFg of Klement et al, (194t)
was confirmed and the Co and Ni analogues were
prepared.
fNrnopucnoN
The primary purpose of this investigation was
to study tle crystal chemistry of wagnerite,
particularly with respect to ions such as Co++,
Ni++, and Cu++ which may prduce
crystals
with unusual colours. However, special empha.sis
was placed oD.Zn** due to its frequent replacement of Mg** in various structures and some
emphasis was plased on Li++, Ca++, V5+, and
Cl-, all of which have a high probability of entering the wagnerite structute, at least in part.
A limited number of trials with charge-coupled
substitutions were made.
replacedby Ca'n, Fe2+,and Mn2+. CaO varies
considerably; the high calcium material perhaps
indicates an alteration toward apatite (Palache
et al. L95I).
Stdctly speaking, the name wagnerite applies
only to the mineral Mgr(PO")F, but this term
has been applied to any substance with the gen'
eral composition AIXOZ. Compounds with the
type formula A;{OA have been systematically
studied in classifying phosphate, arsenate and
vanadate minerals (Richmond 1940). Several
papers have been published involving simple
and charged-coupled substitutions into A, X,
and Z sites (Mourelo 1915; Klement & Gembruck 1941: Klement & Haselbeck 1965; Gorbacheva 1959; Abramsen 1968).
The crystal structure of wagnerite was first
determined by Coda et al. (1967). The most recent report on a new occurrenceof wagnerite in
Colorado (Sheridan et al. 1970) has again confirmed the monoclinic structure and $pacegroup
PL/a. The lattice parameters of wagnerites as
determined by several investigators are shown
in Table 1.
A coordination number of five for magnesium
is unusual, but similar results have been reported for melilite by Christie (1962) and for
tarbuttite by Cocco et al. (1966).
The PDF r-ray pattern for wagnerite is one
which was reported by Henriques (1957) for an
iron-rich mineral from Hallsjobergel Sweden.
Winter (1913) and Berak ef aL (1965) studied
the phase equilibria in the system Mg@O')nMgFa and reported one intermediate compound,
Mga(PO+)fa which melted congruently. Winter claimed a melting point of 1253' and three
polymorphic forrns, and Berak ef al. claimed a
Previous studies
The history and
been reported by
nerite, Mgr(POr)F,
limited quantities,
TABLE1.
NATUMLANDSYN]TETICIIAGNERITES
CELLDIMENS1OI1S.OF
Iatural
properties of wagnerite have
Palache et al. (1951). Wagis monoclinic and occurs in
commonly with Mg partly
ab
Gonyer (ln Richmnd 1940)
Codaet aL. (1967)*
Kraus & l4ussgnug(1935)
.Braltsch (1960)
sher'ldan a|. (1971)
",
SFthetlc (thJs study)
*Abstracted from an M.S. thesis in Ceramic Scieuce
by Keunho Auh at the Pennsylvania State University, December,'1"970,
lo^P0.F
c6:Po:F
Nr:Po;F
cuiPo;F
zniPo;F
346
It.93l
9.6441.007
11.90
il.91
11.929!.005
'It.B7B
1t .918
12.036
12.019
11.944
care
ro8:oe'
12,474
e.441
12.679!.008 1 1 . 9 5 7 ! . 0 0 81 0 8 " 1 8 ! : 9 1
1ogoo7,
9.63
12.51
ro8:20'
9.66
12.69
12.6981.008 9.633r.004 l0B"l?'r3'
12.656
12.675
12.641
12.941
12.672
9.951
9.638
9.688
9.853
9.594
tot.62o^
107.B2:
108.06:
106.94:
107.89"
SOLID SOLUTION
IN THE WAGNERITE
melting point of 1337" and four polymorphic
forms.
ExpsntrvreNrAl-PRocEDURE
STRUCTURE
347
samples were heated for 1-15 hours between
550-1200" in a Globar furnace.
Phase identification
Phase analyses were made primarily with a
Norelco wide-range diffractometer using NiSeveral investigators (Mourelo 1915; Gorba- filtered CuKa radiation in tlre 2A range 10ocheva 1959; Winter 1928; Berak & Tomczak 65'; for compoundscontaining Co**, Ni**, or
radiation wrS ern:
1965) claim to have prepared wagnerite in air. Cu**, Mn-filtered FeKa -'75",
ployed
in
the
2e
range
20o
at a scanning
Two attempts were made by us to prepare the
compound in air, but loss of fluorine and in- rate of 2" /rntn. For more accurate determina7/+o per
complete reaction led us to believe that such tion of d valueso a scanning rate of
minute
and a silicon external standard were
procedureswere unreliable and that synthesisin
sealed platilum tubes was necessaryto assure employed. The midpoint at the half height of
phase-pure wagnerites of reliable stoichiornetry. the peak was taken as the true value of. 20.
A petrographic microscope was used to exIn the first attempt, a stoichiometric mixture
amine
the preparations for impurities and to
(NH.),I{PO4
of MgCOT.MB(OH),,
and MgFr
determine
the optical properties of wagnerites.
was heated from room temperatures to 7L5"
and held for 12 hours. X-ray analysis showed
that Mgb(POa), and unknown phases were
Rpsurts lNo DrscusstoN
present, but no unreacted starting materials.
After remixing in acetone, the product was Characterization ol wagnerite
heated in air at 1020' for 4 hours, giving wagWagnerite was prepared by heating a stoichionerite and rnknown phases. After heating in
air at 113O' for one-half hour, the product con- metric mixture of Mgr(PO"): and MgFz in a
sealed platinum tube at 85O"/3 hrs., 93ool15
sisted of wagnerite and MgO.
A second attempt was made using Mg(POa)a hrs. 1000"/4 hrs., 1040'/4 hrs., and 110O',/4
and MgFz and heating to 1025o for one-half brs. X-ray analysis showed that each heat treathour in an. X-ray analysis gave wagnerite and ment had produced phase-pure wagnerite with
a trace of Mgo@On)2.After L100" for 3 hours, the pattern shown in Table 2, This pattem was
wagnerite and MgO were present. Repeated used as a standard throughout this investigation,
trials showed that about a 95% yield of \yagne- although it did not coincide exactly with the
rite could be expected from this technique, but PDF pattern given by Henriques (1957) for nait was obvious from the two experiments that tural wagnerite.
The claims of Winter (1931) and Berak &
Mg.(POt or MgO would appear and persist in
samples heated at high temperatures in air and Tomazak (1965) for several inversions in wagnerite were not well-documented. Therefore,
that some fluorine was lost.
The technique finally used to prepare all differential thermal analyses were run on wellwagnerites and intermediate solid solutions was crystallized material previously prepared in plato make an orthophosphate intermediate and tinum tubes. Using a Tem-Pres unit, a sample
then react the intermediate with the required of wagnerite and the alumina reference were
fluoride. This technique is particularly success- each sealed in a one-eighth-inch diameter platiful in the laboratory preparation of small sam- num tube, thus preventing fluorine volat:lization
ples of apatites (Kreidler & Hummel 1970) and from the wagnerite. Under these conditions, a
wagnerites. If the ortho-compound intermediate small but distinct heat effect was noted at
was heated between 50&850" for 24 hours, 1255"C during heating and cooling, and melting
usually a single phase ortho-compound was ob- took place at t34Ot:5"C which is in excellent
tained, but complete reaction was not necessary agreement with the L337"C reported by Berek
for the successfulsynthesisof the final single- et al. (L965).
Runs on a du Pont apparatus were made in
phase wagnerite.
Wagnerites were prepared by the use of 0.5 nitrogen and in each casenduring heating and
to 1.0 grarns of ortho-compound and the re- cooling, the inversion was obtained at l255oC.
quired amount of fluoride, mixing at least three All samples from the D.T.A. runs were examtimes for 2O minutes in acetone. The charge ined by r-ray diffraction and found to be wagwas placed in a platinum tube (0.5 cm diameter, nerite. No decomposition had taken place in
2.O cm long) to one-half or two-thirds its length any of the runs.
As a further check on the polymorphism, a
and welded in a D.C. arc. The encapsulated
Synthesis
348
THE
CANADIAN
MINERALOGIST
TABLE2, PARTIALU.MY POI,IDER
DATAFORSYIiII}IETICI,ilAGNERITES*
co2YU4t
Mg2P04F
hkL dcalc
021
o02
121
201
zz0
211
031
T3r
3t0
221
2?n
122
320
202
3 ll
140
for
f4r
fuz
t4l
240
t41
)23
a,ata
{n"as
d"al" dr.u" r/l
a.ty9
4.522
4.450
4.268
4.224 4.232
4.045 4.076
? qla
?
/4
7AE
J.Old
Cu2P04F
NJ2P04F
dcalc
dreas 44
acalc
%eas
Zn?P0AF
//r,
dcil" %uus r/r"
a.t0l
4.532 4.526
4.461
4,408 4.345
4.261
4.251 4.251
8 4.2284.252 2t 4.205 4,200
4 4.040
3.926 3.867
3.857
3.B00
3.809
3.791 3.786
J.OtO
3.615 3.583
7 3 . 5 4 03 . 5 5 9 5 4 3 . 5 1 4 3 . 5 0 6
l 6 3.394
3.368 3.409
5 l 3 . 3 1 03 . 3 i 3 4 6 3.260 3.254
'14
3.244
3.t39 3.132
6 5 3 . 1 2 83 . 1 4 0 7 6 3 . 0 9 5 3 . 0 9 7
C.OJJ
3.541 3.533
3.388 3.439
3.292 3.284
3.244 3,235
3 . 1 2 33 . 1 1 4
3.071
J.UIO
3.054
3.062
3.005 3,002
2.97?
2.985
2.983
2 . 9 7 0 2 . 9 6 5 1 0 0 2 . 9 8 62 . 9 9 61 0 C2 . 9 4 5 2 . 9 3 9
2.827 2.832 40 2 . 8 5 42 . 8 5 0 9 l 2 . 8 3 5 2 . 8 3 6
2.822
2.835
2.776 2.786
2.816 2.802 21 2 . 8 1 8
2.803
2.767
2.773
2.745 2.740
2.764
2.7852.794 26 2 . 74 6
2 . 7 4 7 2 . 7 4 8 28 2 . 7 7 92 . 7 7 1 5 8 2 . 7 2 2 2 . 7 2 3
n
14 4 . 6 1 04 . 6 0 2
1 2 4.498
14 4 . 2 9 2
22 4.2514.176
15 4 . 0 7 24 . 0 4 4
3.890
l 8 ? qlr
l6 J.OJ+ J.OZI
6 5 3 . 5 7 13 . 5 9 4
l 4 3 . 5 4 43 . 5 0 6
3l 3 . 3 2 43 . 3 1 8
2 1 3.265
5 8 3 . 1 4 53 . 1 5 0
? no6 1 no6
2a4.490
4.474
4 . 2 9 04 , 3 0 5
21 4.2474.256
2t 4.065
J.Utv
5.d9/
3.807
JJ
42 J . t c o J . l o u
6i
6i 3.403
3 ( 3 . 3 t 73 . 3 1 7
3 . 1 8 43 . 1 8 7
6 4 3 . 1 4 43 . 1 4 2 7 l
ir 3 . 0 8 63 , 0 8 3
t
l4 3.089
J.UOO
2.9812.964 44 2.993
1 0 02 . 9 1 12 . 9 1 6 9 C2.9862.990 '10(
9!
8 8 2 . 8 3 12 . 8 2 8 1 0 C2 . 8 5 02 . 8 5 r
1 4 2.864
2.841
2.U0
2.829
23 2.800
2.779
2.797
2.780
7 2 2.7732.807 5l 2 . 7 7 62 . 7 7 7 71
are
wagnerite solid solution containing a L5 mole /s
Cobalt substitution
substitution of Co++ for Mg++ inverted. at
Using (Mgi-J.{t)"(POn)r, MgFr, and NiFz,
L235"C in the Tem,Pres unit when in a sealed
'l-.2,
sampless6af4ining 0.6,
I.8, 2.4, 3.0, and
platinum tube.
(CorPzOeFz)
moles of Co+ + were heated at
It was concluded that waperite undergoes 4.O
only one inversion, at I255"C, prior to melting 850'/3 hrs. to produce a complete series of
solid solutions, yielding a variation in peak posiat L34O.C.
tions as shown in Figure 1. Heating at 9M"/6
hrs. gave phase-pure solid solutions, but at
1000"/24 hrs., wagnerite and unidentified
-a
(Ms,co)a%oBFz(85ooc)
phases were present.
--_-+
(Nlg,Ni)4
%OsFz ilO8ooc)
---{
Attempts to prepare cobalt wagnerite or in0ro,cu)aPr8F2 {87ec)
-'-'-€
([as,zn)aPz%F2
(goo"c)
termediate members of the solid solution series
by heating a mixture of Mg,Co),(POn),, CoF,
and MgFe, or 3MgCO;.M9(OIDz,CoCOs,(NIL)2.
HPOe, CoFz,and MgFz at7l5"/ 12 hrs., LOAO'/
4 hrs., or 1,18O"
//z hr. in air were unsuccessful.
Nickel substitution
o
o
u
z
3
s
o
Y
U
Using (Mgr*Ni,)a(POn)a MgF , and NiFr,
samplescontaining 0.6, L.2, L.8, 2.4, 3.0, and
4.0 (NioPrOrFg)moles of Ni++ were heated at
1080'/3 hrs. to produce a complete series of
solid solutions, yielding a variation in peak positions as shown in Figure 1. The compositions
were stable if heated to l2OO"/ 4 hrs., but higher
temperatures or longer times produced wagnerite and secondaryphases.Synthesisatttempts
in air were not successful. The r-ray data for
NifsO8Fz are shown in Table 2.
Copper substitution
MqP2O6F2
N
60
80
too
MOLE%+
Frc. 1, Variatio\ of W yalues as a function of com.
position for wagnerite solid solutions.
Using (Mgr*Cu,), (POo)r, MgFr, and C\rF%
samples containing (Cu4PrOBFr)moles of Cu+ +
were heated at 87O"/l hr. to produce a complete series of solid solutions, yielding a variation in peak positions as sho,wn in Figure 1.
SOLID
SOLUTION
IN
THE
WAGNERITE
Heating at 850'/1.5 hrs. producedpure phases,
but at IC00" /4 hrs. wagnerite and unknown
phases were present. The bright green colour
became more intense as the amount of copper
was increased, but as the samples were heated
at higher temperatures or longer times than the
optimum, the colour changed to light brownish
gay, indicating reduction of Cu*+ to Cu+. Xray diffraction data for CuaPzOeF2
are shown in
Table 2.
Zinc substitution
Using (Mgr"Zn")u (POo)r, MgFz, and ZnFz,
samplescontaining 0.6, L.2, L.8, 2.4, 3.O, and
4.O (Zn&zOaFr)moles of Zn++ were heated at
8OO"/6 hrs. to produce a complete series of
solid solutions, yielding a variation of peak
positions as shown in Figure 1. The zinc substitutions contract the structure in a manner almost identical with cobalt" whereas Ni** and
Cu** expand the structure. X-ray diffraction
data for Zn&uO"F, are shown in Table 2.
Heating at 11AO"/L hr. in sealed tubes produced Mg(POa)z and Zns(POo)"as major phases
and synthesisin air was not successful.
-{
--{
--Q
349
(Zn,Co)4 P2OaF2 (82O"C)
(zt1,Ni )4 P2o8 F2 ( I loo" c )
(Zn, Cu )a P2OBF2 (8OOoC)
z
E
@
u
c
zn4P2oaFe
Substitution in the zinc site ol ZnaPzOeFz
and
ZnsMgPzOaFz
STRUCTURE
MSLE%_
Frq. 2. Variation ot 20 values as a function of composition for ZnfzOsF, solid solutions.
Using (7.n1-"Qs")gPOr,(Znr-"Nt)a(POn)r, Zrttn
Cu")r(POa)a MgFa, ZnFn, CoFg, NiFr, and
-+
(zn,co)s
lioaosFa(smoc)
CuFz, molar substitutions of Co+*, Ni**o and
---'4
(Zn,Ni)r
(l@ot a 8@'C)
t'lgP2o6F2
Cu++ were made in ZnnPzOaF,and ZnuMgPz,
(Zn,Cu)3MqP2Os F2 ( 9OOoC)
OaFr at levels of 1.0, 1.6, 2.2, 2.8, 3.4, and
0.6, I.2, 1.8, and 2.4, respectively. Heat treatments were at 820"/L.5 hrs., 1100'/1.5 hrs.,
and 82O'l1.5 hrs. for Co++, Ni++, and Cu+*
in ZnaPzOaFaand at 800 - 900"/2-3 hrs., 80OlOC0"/2-3 hrs. and 9W"/Z hrs. for Co++, N
Ni++, and Cu++ in ZnsMgPzOeFr!,
respectively. 2
These solid solutions were much more sensi- E
tive to temperature and time of heat treatment d
than those previously described and carefuI gY
choice of both variables had to be made in
order to avoid the presenceof unreacted starting materials at low temperatures or secondary
phases at temperatures higher than the optimum. Tho vadations of 2d values as a function of composition are shown in Figures 2
and 3. The maxima and non-linearity in the
curves indicate that some interesting packing
effects are in operation as the transition metal
ions substitute for Zn++. The expanding effect
of Cu++ is still evrdent, but much modified
from the linear behavior shown in Figure 1.
The compositions which showed maximum
deviation from linearity in Figures 2 an'd 3 Frc. 3. Variatiot of 20 valuesas a funstion of comgave .r-ray peak intensities which were conposition for ZneMgPgOeF,solid solutions.
@
d
350
THE CANADIAN MINERALOGIST
site. He gave no evidence to substantiate his
claim. The temperature and pressure conditions under which fluorine spodiosite can be
synthesized apparently have not yet been deCalcium substitution
but the chlorine analogue, (CaaP:Or
termined,
Ca+ + was found to be soluble in MgoPzO.F, Clr), has been prepared by heating in air by
and CooPzOeF, but not io Ni4PIOBF2 and many investigators.
CufzOsFz. Using the presence of a secondary
phase as an indicator, the limits of solubility Lithium substitution
shown in Figure 4 were established.Large expansions of the structure were obseryed. Heat
If lithium is substituted for magnesium so
as to preserye a total of four atoms in the z4
treatmentswere at 93O'/L5 hrs. and 830'/4
hrs. in the Mg and Co senes, respectively. site, halide vacancies would be required to
The work on synthesis and luminescence of preserye a charge balance. If as many as two
wagnerites by Gorbacheva (1959) is suspect atoms of lithium could be substituted in the
because all heat treatments were made in air at magnesium site, this would produce the comtemperatures where it was likely that ortho- pletely halide-vacant wagnerite, Li:MgrPrOrnr,
similar to the well-known completely halide'
phosphates would be formed.
Subsequent studies of the Mn++-activated vacant lead apatites. If half the halide sites
luminescence of the so-called wagnerites of were vacant, the formula would be LiMgJr
Gorbacheva by B. Bacik (1970) have shown O.Ftr. It was found that only 0.6 mole of Li'
his original work to be unreliable. His claim was soluble in Mg++, Co*+, and Ni** wagnefor the preparation of Ca"PrOsFzBS & wagne- rites, giving the general formula, Li*o.oMgg.arite is especially doubtful, since this is the PzOoFr.alo.s.Heat treatments were at 93A"/ 4
composition ascribed to the mineral spodio' hrs., 620'/3 hrs., and 620'13 hrs. for Mg**,
Co* +, and Ni* *, respectivelY,
siderably different from the end members, indicating substantial distortion of the structure.
Vanadium and chlorine substitutions
J
-+
{uo,co)oe. o.F2 (93OoC)(usedcuKq Rqdiotion)
(Co,Co)aP2O8F2
(g3o"c) (us€dFe(x Rodidtion)
Y'
2.904
o 3ZBO
N
z
F
6
q
U
37.70
)
f
P
'"i"7
2940
About 40 mole /o V5+ can replace P5+ in
MgnPrOeFa, CooPrOeFr, Ni"PrOeFr, and 7-na'
PrOrFr, using heat treatments of 1O50o/4 hrs.
for MgnPrOsFza\d 7O0o/L2 hrs. for the other
three compositions.
Klement & Haselbeck (1965) reported the
synthesis of MgnPrOeClzand Co4PrOeClr,but
all attempts to duplicate these results by heating in sealed tubes or air failed. The use of
an excess of NHaCI in the starting mixture was
also unsuccessful. Poorly-crystallized CunPlor
Cls and ZnqPzOeClzw€re prepared by heating
in air at 3OO' and 180' for two hours. Traces
of unknown phases were present in both
preparations and the zinc compound melted
at 30Oo. It is possible, of course, that partial
substitution of Ct for F- could be made in
most of the wagnerites so far mentioned, but
this was beyond the scope of the present work.
Charge-coupled substitutions
2.959
30.t0
40
I0
20
30
MO-E "/. Co4PaOeFz ---------*
Frc. 4. Solubility of Ca in MeaPzOsFz and CoaPgOr
Fq.
Klement et al. (1941) had claimed the
preparation of LiMgPSOaFz and NaMgPSOeF,
as wagnerites and this was confirmed in the pre'
sent work by heat treating MgSOa, MgFz and
LiMgPOe or NaMgPOa at gAOo/2 hrs. At 100O'
/2hrs., the compoundsbegin to convert to other
phases. The analogues LiCo.PSOrFr, NaCogPSOoF:, Lil{isPSOeF, and NaNisPSOeFe were
SOLID
SOL{JTION
IN
THE
WAGNERITE
STR{JCTURE
35L
prepared by heating at 9OO"/2 hrs,, 70Oo/2 Bsner, J. & Totuczer, I. (1965): Phase equilibria in
the system MgO-PzOs-MgFz. Rocznikl, Chem, 39,
hrs., 900"/2 hrs., and 9Q0"/2 hrs., respec519-525.
tively. The sodium-nickel compound was poorBRArrscH, O. (1960): Borates and phosphates in tbe
ly crystallized, but phase-pure.
Zechstein salt of South Hannover. Fortschr, Mineral. 38, 190-191.
Suvrvranv
CnrrsnE, O. H. J. (1962): On sub-solidus relations
of silicates III. A contribution to the chemistry
Conditions for the synthesis of phase-pure
of melilites. Norsk. Geol. Tids. 42, l-29.
wagnerite have been established. Wagnerite
Cocco, F. FeNreNr, L. & Zntvzr, P. (1966): Tbe
undergoes only one reversible polymorphic incrystal structure of tarbuttite. Zeits. Krist. L2?,
version at 1255", contrary to the several in
versions reported previously. Analogous Co++,
Ni**, Cu**, and Zn++ wagneritesform complete series of solid solutions with magnesium
wagnerite, and Co++, Ni++ and Cu+* form
complete solid solution series with zinc wagnerite. Exploratory work on the substitution of
Ca*+, Li*, Vt+ and Cf in the new wagnerites
showed. that partial substitution was possible
for Ca++, Li+ and V5+ in some of the preparations, but highly specific behavior was evident. Complete substitution of Cl- for F- was
possible only under carefully controlled conditions in copper and zinc wagnerites. New
charge-coupled cobalt and nickel wagnerites
were prepared.
32r-329.
Cooe,A., Grusrernru, G. & Teorm,C. (1967):
The crystal structure of wagnerite. Atti, Accad.
Naz. Lincei. Rend. Cl. Sci. Fis. Mat. Natur. 43,
2t2-224.
Fucns, J. N. (1821): Phosphorsaurer talk, Wagnerjte. l. Chemie Phys. 35, 269-277.
Gonreqreve, N. A. (1959): Fluorophosphate phosphors similar to mineral wagnerite. Izvest. Akad.
Nauk USSR (Ser. Fiz.) 28, 131,0-1313.
HENRreuEs,A. (1957): An iron-rich waperite, forfrom Hallsjoberget
mally named talktriphite
(Horrsjoberget), Sweden. Arkiv. Mineral, Geol.
2, t49-ts2.
F. (1941):Experiments
Kr.rurNl R. & GBnonucK,
on isomorphism of the wagnerite group, Naturwiss. 26, 3Ol-302.
& Hesar,nrcr; H. (1965): Apatites and
wagnerites of divalent metals. Z. Anorg. Chem,
AcxNowr,stcEMENTs
336, Lt4-128.
The authors are grateful to Dr. D. K. Smith Kneus, O. & MusscNuc, F. (1938): Gitterkonstanfor use of the facilities which provided. the inten und Raumgrupp: von Wagnerite Mg(MgF)
dexed powder patterns (done under a grant-inPOa. Naturwiss. 26, 801-802.
aid from the Joint Committee on Powder Knr,mr-rn, E. R. & HurvrrvrEr.,F. A. (1970): Thc
crystal chemistry of apatite: structure fields of
Diffraction Standards) and for his critical apfluor- and chlorapatite. Amer. Mineral. 55, 170praisal of the diffraction data. Thanks also
184.
are due to Dr. R. A. McCauley for his translation of the paper by Gorbacheva, to Miss Mounsro, J. R. (1915): Methods of mineral synthesis. Rev. Gen. Sci. 26, 394-403.
M. Schulan for compilation of x-ray data and
to Mr. TV. Harding, Miss S. Kemberling and PALAcHE,C., Bnnuen, H. & Fnorprr., C. (1951):
The System of Mineralogy, vol. If, 7th edit.,
Miss Mary Hoffman for D.T.A. runs on
John Wiley & Sons, New York.
wagnerite compositions. This work was supported in part by the Ferro Corp., Cleveland, RrcHMoND, $/. E. (1940): Crystal chemistry of tfte
phosphates, arsenates, and vanadates of the type
Ohio.
ArXOtZ, Amer. Mlneral. 25, 44L-479.
SnrRroeN, D. M., Mensn, S. P., Mnose, M. E. &
Tevr.on, R. B. (1971): Wagnerite from Santa Fe
RenERrNcrs
Mountains, Colorado: a new occurrete. Can,
Mineral, L0, 9L9 (abstr.).
Arrausrr, W. (1968): Color in the Wagnerite
Structure. B.S. Thesis, Ceramic Sci. Sbction, Ma- WtNttn, H. (1913): Dissert., Leipzrg. ln Internat.
terials Sci. Dept., Penna. State Univ.
Crltical Tables 4 (1928).
Becrr, B. (1970): Synthesis ntd Luminescenceof
Wagnerite. Unpub. rept., Ceramic Sci. Section,
Material Sci. Dept.,Penna.StateUmv.
Manuscript received tanuary 1972, emended tanuary 1974,