Laboratory Formation of Taranakite — Liu, Sherman, and Swindale 
499 
and 95% ethanol by decantation (Wada, 1959) 
to remove excess phosphate. The finer fractions 
of the solid phase were collected on the filter 
paper. The coarser fractions were left in the 
glass container. Both fractions were then air- 
dried. 
Potassium phosphate solutions having con- 
centrations ranging from 0.2 molar to 0.05 
molar, or from 0.6 molar to 1 molar, with 
initial pH values of 2-4, were also used, but 
crystalline reaction products did not form. 
A taranakite sample was synthesized for com- 
parison; the synthesis was made as follows: 
20 ml of 65% phosphoric acid were mixed with 
an aluminum chloride solution containing 2.65 
g of aluminum chloride. The mixture of 
phosphoric acid and aluminum chloride was 
filtered and diluted to 50 ml, and an aliquot 
was taken in a pyrex beaker and neutralized 
with 10% potassium hydroxide to a pH be- 
tween 3.4 and 5.0. The beaker containing the 
flocculent precipitate was covered with a watch 
glass and kept in a 50 °C water bath. The 
crystalline precipitate which formed was sepa- 
rated by filtration, washed with distilled water, 
and air-dried. 
Characterization of Phospho-reaction Products 
by X-ray Analysis 
The reaction products were identified by 
X-ray diffraction, using a Norelco X-ray dif- 
fractometer with copper Ka radiation. The 
X-ray diffraction patterns obtained throughout 
this study were of powder samples. 
The effect of heating on the crystalline sub- 
stance resulting from the soil-phosphate system 
and on the synthesized taranakite was exam- 
ined. Both were subjected to a range of tem- 
peratures from 60 °C to 150°C. The products 
resulting from heat treatment were identified 
by X-ray diffraction. 
X-ray diffraction patterns of the phospho- 
reaction products prepared by treating Akaka 
soil with solutions containing potassium phos- 
phate, ranging from 0.2 to 0.6 molar at pH 2, 
showed these reaction products to be taranakite. 
D-spacings for taranakite and heated taranakite 
obtained in these experiments and by Haseman 
et al. (1950) are gathered in Table 2. The 
prominent peaks of natural taranakites from 
Sugarloaves, Taranaki, New Zealand, and from 
Pig Hole Cave, Giles County, Virginia, are at 
16.2, 7.6, 3.82 A, and at 15.49, 7.82, 3.79 and 
3.12 A, respectively. The 16.2 and 15.49 A 
peaks are the strongest. The synthesized tarana- 
kite produced in this experiment, and those of 
Haseman et al. (1950), showed the strongest 
peak at 15.7 A; and strong peaks at 7.89, 3.81, 
3.14 and at 7.35, 3.79, 3.13 A, respectively. 
Except for the 7.89 A peak given by the phos- 
pho-reaction product obtained from Akaka soil, 
the prominent peaks of natural and synthesized 
taranakite are similar to the reaction products 
obtained from the Akaka soil. Birrell (1961) 
identified a taranakite by treating a Tirau ash- 
clay fraction from New Zealand with potassium 
phosphate solution. The X-ray diffraction pat- 
tern of the taranakite studied by Birrell showed 
prominent peaks at 15.60, 3.81, and 3.13 A, 
and is very similar to that of the phospho- 
reaction products obtained from the Akaka soil. 
The reaction products resulting from soil- 
phosphate systems, and the synthesized tara- 
nakite, were heated to different temperatures 
in order to examine the dehydration products, 
as well as the effect of heating on crystallinity. 
The dehydration products resulting from the 
soil-phosphate systems and the synthesized 
taranakite were similar with d-spacings of 13.6, 
6.8, 5.5, and 3.4 A. This substance is a com- 
pound like Product B as synthesized by Hase- 
man et al. (1950). Both the reaction product 
resulting from soil-phosphate systems and the 
synthesized taranakite became amorphous above 
130°C 
Characterization of Phospho-Reaction Products 
by Chemical Analysis 
The evaluation of K/PQ 4 and A1/P0 4 ratios 
for the reaction product was difficult because of 
contamination of the reaction product by soil 
particles. Therefore, a factorial experimental 
design (Snedecor, 1956) with four replications 
at three levels of soil was prepared, using 5, 
10, and 15 g samples of wet soil with a 0.4 
molar potassium phosphate solution at pH 2. 
Phosphate was estimated by chlorostannous- 
reduced molybdophosphoric blue color in a 
hydrochloric acid system, following fusion of 
the sample with sodium carbonate, and decom- 
