International Journal of Pharmaceutics, 42 (1988) 171-180 171

Elsevier

IJP 01423

An examination of the physical properties

of pure magnesium stearate

K.D. Ertel and J.T. Carstensen

School of Pharmacy, University of Wisconsin, Madison, WI 53706 (U.S.A.)

(Received 16 April 1987)

(Modified version received 12 August 1987)
(Accepted 7 September 1987)

Key words: Magnesium stearate; Lubricant; Pseudo-polymorphism; Polymorphism; Rehydration; Hydrate

Summary

The effect of various preparation conditions on the form of magnesium stearate prepared by aqueous precipitation was examined.
None of the preparation variables had any apparent effect on the crystal habit of magnesium stearate; however, the preparation
conditions did affect the amount of moisture associated with the crystals. Heating the magnesium stearate at 105 o C not only resulted
in a loss of water of hydration from the crystals, but also a disruption of the crystals’ 3-dimensional lattice structure. Both of these
effects could be reversed by storing the dried crystals under appropriate relative humidity conditions.

Introduction Shah, 1977; Iranloye, 1978; Billany, 1982). Conse-

quently, it is desirable to optimize the amount of
Magnesium stearate is widely used as an excipi- magnesium stearate used in a formulation, i.e. to
ent in the production of compressed tablets, due use an amount which adequately lubricates the
mainly to its ability to decrease friction between formulation but minimizes the adverse effects on
the tablet’s surface and die wall during the ejec- the finished tablets. Unfortunately, the optimiza-
tion process. The compound is usually effective in tion process is complicated by the fact that com-
low concentrations; however, some care must be mercial magnesium stearates exhibit significant
excercised when preparing formulations with mag- batch-to-batch variation in their ability to func-
nesium stearate due to the compound’s hydro- tion as lubricants (Hanssen, 1970; Miiller, 1977a;
phobic nature, which has been shown to adversely Steffens, 1982; Rubinstein, 1983; Colombo, 1984;
affect tablet properties such as hardness (Strick- HGlzer, 1984).
land, 1956; Shah, 1977; Lerk, 1977a; De Boer, Although attempts have been made to identify
1978; Ragnarsson, 1979), disintegration time the factor(s) responsible for the batch-to-batch
(Strickland, 1956; Shah, 1977; Ragnarsson, 1979), variation in the lubricant properties of magnesium
and dissolution rate (Levy, 1963; Lerk, 1977b; stearate by examining commercial samples, these
studies are complicated by the fact that commer-
cial magnesium stearate is not an exact chemical
Correspondence: K.D. Ertel, Present address: Merrell Dow entity. For this reason, a number of workers have
Research Institute, Cincinnati, OH 45215, U.S.A. chosen to use pure magnesium stearate, which

0378-5173/88/$03.50 0 1988 Elsevier Science Publishers B.V. (Biomedical Division)

172

may be prepared by precipitation from an aque- Preparation of magnesium stearate

ous medium according to the reaction: All magnesium stearates used in this work were
prepared by precipitation from an aqueous
2 CH,(CH,),,COOX + M&l, medium. After dispersing approximately 0.01 mol
of stearic acid in 600 ml of water (previously
-+ [CH,(CH,),,COO],Mg + 2 XC1 heated to 90 ’ C), the sodium or ammonium soap
of the fatty acid was formed by dropwise addition
where X is a cation such as NH: or Na+. Both
of saturated sodium hydroxide solution or am-
plate-shaped dihydrate and needle-shaped trihy-
monium hydroxide, respectively. The addition of
drate forms of magnesium stearate have been pre-
base was continued until the pH of the soap
pared by this method, the hydration state being a
dispersion was in the range 7-11. In most cases,
function of the pH of the precipitation medium
precipitation was carried out at pH 9. An amount
and the cooling rate (Mtiller, 1977a-c, 1982;
of magnesium chloride slightly in excess of that
Miller, 1982, 1985). The hydrates have different
required for stoichiometric reaction with the fatty
crystal habits and, more importantly, belong to
acid soap was dissolved in 100 ml of water and the
different crystal systems (Mtiller, 1977~). The latter
resulting solution was added dropwise, with stir-
is a result of changes in the angle of inclination of
ring, to the soap dispersion over a 20-min period.
the hydrocarbon chains relative to the plane of the
The mixture was cooled to room temperature over
magnesium atom head groups brought about by
a period of 3-12 h, after which time the precipi-
the incorporation of water into or the loss of water
tate was isolated by filtration, transferred to a
from the crystal lattice. An anhydrous form of
Soxhlet extraction apparatus, and washed for 24-h
magnesium stearate, prepared by drying hydrous
periods with acetone and water to remove any
magnesium stearate under appropriate conditions,
unreacted fatty acid and magnesium chloride. The
has also been described. The hydrocarbon chains
washed powder was air-dried and stored under
of the anhydrate are packed in a manner different
ambient conditions.
from that of either of the previously mentioned
Anhydrous magnesium stearate samples were
hydrates (Vold, 1949; Mtiller, 1977a, c).
prepared immediately before use by drying hy-
drous magnesium stearate to constant weight at
Materials and Methods 105 o c.

Materials Hydration behavior

The purity of the stearic acid (99 + %, Aldrich To examine the hydration behavior of mag-
Chemical Company, Milwaukee, WI) was verified nesium stearate, desiccators containing saturated
chromatographically. Magnesium chloride (ACS solutions of the following salts were prepared
grade, Drake Brothers, Menomonee Falls, WI), (Nyqvist 1983; Weast, 1977): LiCl(11.3%), MgCl,
acetone (AR grade, Mallinckrodt Inc., St. Louis, (32.8%) K&O, (43.0%), Mg(NO,), (52.8%) KI
MO), ethanol (absolute, Midwest Solvents Com- (68.9%), and KBr (80.9%). The desiccators were
pany of Illinois, Perkin, IL), and ammonium hy- stored at 25 + lo C to maintain the relative humid-
droxide (ACS grade, J.T. Baker Chemical Com- ities shown in parentheses. In addition, a desicca-
pany, Phillipsburg, NJ) were used as received from tor containing phosphorous pentoxide was
the suppliers. Potassium bromide (spectroscopic prepared to produce a relative humidity of 0%.
grade, Aldrich) was dried for 2 h at 110°C before
use. A saturated solution of sodium hydroxide was Thermal properties
prepared from sodium hydroxide pellets (reagent The thermal properties of pure magnesium
grade, Amend Drug and Chemical Co., Irvington stearate samples were determined using a thermal
NJ). Water used throughout was double-distilled analysis system (Series 7, Perkin-Elmer Corp.,
from an alkaline permanganate solution in an Ridgefield, CT) which included a differential
all-glass still. scanning calorimeter (DSC) and a thermogravi-
 173

metric analyzer (TGA). DSC examinations were in these factors had any effect on the crystal habit of
most cases carried out over a temperature range of the magnesium stearates produced in the present
40 o C-150 o C at a heating rate of 2’ C/mm, al- work, as all of the samples consisted of plate-
though in some cases samples were heated to shaped crystals, i.e. the needle-shaped crystals de-
temperatures as high as 200 ’ C at a heating rate of scribed in the literature could not be isolated. The
lO’C/min. TGA examinations were carried out medium pH did have an effect on the amount of
over a temperature range from 50°C to 130” C at water associated with the crystals, however.
a heating rate of 5 o C/mm. Batches precipitated at pH 9 were shown by ther-
The appearance of the magnesium stearate dur- mal analysis to contain an amount of bound water
ing heating was observed under magnification corresponding to a dihydrate (form A), while
using a polarizing microscope (Optikot-Pal, Nikon, batches precipitated at the pH extremes (i.e. 7 and
Japan) equipped with a hot-stage (Model FP 82, 11) contained considerably less bound water (Ta-
Mettler Instrument Corp., Hightstown, NJ). A ble I). The decreased amount of moisture associ-
heating rate of 2” C/mm was maintained by a ated with the latter batches may be due to irregu-
central processor (Model FP 80, Mettler). lar incorporation of water into the magnesium
stearate crystal lattice during precipitation (Miller,
Infrared spectroscopy 1982) or to the formation of a mixture of mag-
The infrared absorption spectra of stearic acid nesium stearate dihydrate and a lower hydration
and of hydrous and anhydrous magnesium stearate state (Miller, 1985).
samples were obtained using an infrared spectro- Earlier workers have reported that it is possible
photometer (Model 599B, Perkin-Elmer Corp., to prepare magnesium stearate trihydrate by aque-
Norwalk, CT). Ca. 1 mg of the powder to be ous precipitation. Despite the wide variety of
examined was ground with 100 mg of dry potas- precipitation conditions employed in the present
sium bromide, compressed into a pellet, and study, the trihydrate could not be isolated, al-
scanned from 200-4000 cm-‘. though several batches of magnesium stearate
precipitated at pH 9 did contain slightly more
X-ray analysis
water than the amount expected for a dihydrate
Diffraction patterns for magnesium stearate
(e.g. 2.4-hydrate). Storage of these powders over
were obtained using a powder diffractometer
phosphorous pentoxide, however, decreased their
(Phillips Electronics Inc., Mount Vernon, NY).
moisture content to that of a dihydrate (Fig. l),
Samples were irradiated with X-rays from a copper
indicating that the excess water was present as
target (Cu kar radiation, h = 0.15418 nm). Two
adsorbed surface moisture.
types of samples mounts were used: random
An anhydrous form of magnesium stearate
powder mounts, prepared using the sample holder
(form B) was prepared by drying the dihydrate to
provided with the instrument, and ordered powder
mounts, prepared by dropping an ethanolic sus-
pension of magnesium stearate crystals onto a TABLE 1
clean glass microscope slide and allowing the Effect of precipitation pH on the moisture contents of various
solvent to evaporate. Diffraction patterns from Iaboratoty-prepared magnesium stearates
both types of mounts were measured from 20 = 5 o
to 25 O, where 0 is the Bragg angle. Precipitation pH Loss on drying (weight W)

9 5.5650.17 (n = 4)
9 5.71fO.O8(n=8)
Results and Discussion 9 5.22*0.11 (n = 3)
9 5.54k0.21 (n = 4)
As noted in the introduction, medium pH and 7 2.06 + 0.01 (n = 3)
11 3.42+0.31 (n = 8)
cooling rate are thought to be the primary factors
which determine the form of magnesium stearate Values are mean + S.E.M., n denotes the number of determina-
prepared by aqueous precipitation. Neither of tions. The theoretical weight loss for a dihydrate is 5.74%
174

2.4 HYDRATE OVER P,Os 11.3% RELATIVE HUMIDITY (DRIED)

Fig. 1. Weight loss of magnesium stearate form A (2.4-hydrate) Fig. 2. Moisture uptake of magnesium stearate anhydrate
when stored over phosphorous pentoxide. Error bars represent (form B) when stored under 11.3% relative humidity condi-
95% confidence limits. tions Error bars represent 95% confidence limits.

constant weight at 105 o C. This treatment resulted relative humidity was less than 50%, as evidenced
in a loss of moisture only; the plate-shaped ap- by the rapid and continuous weight loss observed
pearance of the original dihydrate was retained. during thermogravimetric analysis (Fig. 4). This
The loss of moisture was, however, accompanied type of moisture sorption has previously been
by the development of striations in the crystals described for anhydrous magnesium stearate
and a darkening of the crystals’ appearance when (Miiller, 1982). When the relative humidity ex-
viewed under crossed polars, indicating a loss of ceeded 50%, however, the anhydrate rehydrated to
their anisotropic property, as has been described form a trihydrate (form C). This phenomenon has
previously (Miller, 1985). These changes were most not been previously described for anhydrous mag-
likely due to desolvation rather than to a polymor- nesium stearate, although other anhydrous metal
phic change in the crystals. soaps are known to rehydrate under appropriate
Magnesium stearate dihydrate exhibited no conditions. The rehydration process was accompa-
tendency to sorb moisture when stored over the
various saturated salt solutions described earlier.
52.8% RELATIVE HUMIDITY (DRIED)
This is not surprising, since magnesium stearate is
classified as being non-hygroscopic (Callahan,
1982), although the compound apparently will sorb
moisture when stored under relative humidities
greater than approximately 85% (Callahan, 1982;
Hiilzer, 1985; Handbook of Pharmaceutical Ex-
cipients). Furthermore, except for the specific cases
mentioned above, the dihydrate exhibited no tend-
ency to lose moisture when stored over phosphor-
ous pentoxide.
Unlike the dihydrate, the anhydrous form of
magnesium stearate did exhibit a tendency to sorb
moisture when stored over the various saturated Fig. 3. Moisture uptake of magnesium stearate anhydrate
salt solutions (Figs. 2 and 3). This moisture was (form B) when stored under 52.8% relative humidity condi-
present as adsorbed surface moisture when the tions. Error bars represent 95% confidence limits.
 175

TEMPERATURE (‘C) TEMPERATURE 1°C)

Fig. 4. Typical % weight vs temperature curve (TGA) for Fig. 5. Typical % weight vs temperature curve (TGA) for
magnesium stearate anhydrate (form B) stored under relative magnesium stearate dihydrate (form A).
humidities less than 50%. The example shown is from a sample
stored under 11.3% relative humidity.
characterized by DSC. Thermograms of both the
dihydrate (Fig. 7) and trihydrate (Fig. 8) forms of
nied by a significant increase in the crystal long magnesium stearate exhibited two thermal events,
spacing of the magnesium stearate (Table 2), indi- although the peak temperatures associated with
cating that a change in the packing of the hydro- these events were shifted to lower values for the
carbon chains had occurred. Once formed, the latter form (Table 3). No thermal events were
trihydrate was stable at 25’ C when stored under observed at temperatures greater than 130°C for
relative humidity conditions less than those any of the samples. Previous workers have as-
required for its formation, i.e. less than 50%. The signed the low temperature endotherm to the loss
anhydrate could be regenerated, however, by dry- of bound water from the crystals and the high
ing the trihydrate at 105 o C. temperature endotherm to a melting phenomenon
Thermogravimetric analysis of the dihydrate (Mtiller, 1977a, c; Miller, 1985). These assign-
and trihydrate forms of magnesium stearate ments were confirmed in the present work by
showed that the waters of hydration within a given TGA and hot-stage microscopy. As shown in Table
hydration state were bound equivalently, as evi- 3, the enthalpic change associated with the loss of
denced by the lack of shoulders in their weight bound water from the dihydrate was significantly
loss curves (Figs. 5 and 6). The thermal behavior greater than that associated with the loss of bound
of all 3 forms of magnesium stearate was further

TABLE 2
Crystal long spacing (do,,) of magnesium stearate anhydrate
(form B) as a function of relative humidity

Relative humidity (%) d,, (nm)

0.0 5.023+0.005 (n =lO)
11.3 4.999 + 0.006 (n = 10)
32.8 5.034 f 0.006 (n = 16)
43.0 5.056rtO.005 (n = 6)
52.8 5.270*0.003 (n =16)
68.9 5.271+ 0.003 (n = 13)
80.9 5.304*0.003(n =18) TEMPERATURE (“C)

Values are mean f S.E.M., n denotes the number of determina- Fig. 6. Typical & weight vs temperature curve (TGA) for
tions. magnesium stearate trihydrate (form C).
176

TABLE 3
Thermalpropertresof the various forms of magnesium stearate as
determined by DSC

Form T, (“C) AH, (J/g) Tz (“C) AH, (J/g)

Aa 100.1 229 122.5 58.6
Bb _ _ 125.1 59.7
C” 82.9 146 116.7 35.1

a Heating rate 2” C/mm

50 70 90 110 130
h Heating rate 10 ‘C/mm.
TEMPERATURE PC)

Fig. 7. Typical thermogram (DSC) produced by heating mag-

nesium stearate dihydrate (form A) at a rate of 2” C/mm. powder diffraction using randomly oriented sam-
ples (Fig. 10). Of particular interest is the region
near 20 = 21”. The diffractogram of the dihydrate
water from the trihydrate, indicating that the
exhibited several distinct peaks in this region,
bonds holding the water in the two hydration
while in the case of the anhydrate, these peaks
states were of different strengths, i.e. water was
were replaced by a single broad peak. This type of
bound more tightly in the dihydrate than in the
peak is known as a halo, and is indicative of a
trihydrate. This finding is consistent with what has
structure in which the magnesium atoms of mag-
been reported previously for magnesium stearate
nesium stearate are arranged in irregularly spaced
dihydrate and trihydrate (Miiller, 1977b). The
parallel planes, i.e. the three-dimensional structure
thermogram obtained from the anhydrous form of
of the crystal lattice has been disrupted (Vold,
magnesium stearate (Fig. 9) exhibited a single
1949). The diffractogram of the trihydrate again
endothermic transition, that associated with the
displays several distinct peaks in the region near
melting of the crystals. For the particular case
20 = 21°, indicating that the three-dimensional
shown, the lack of an endotherm corresponding to
lattice structure has been restored by rehydrating
the loss of bound moisture from the crystals pro-
the crystals.
vides further evidence that moisture sorbed when
Crystal long spacings for the various forms of
the anhydrate was stored under relative humidities
magnesium stearate were calculated from data ob-
less than 50% existed as adsorbed surface moisture
tained from diffractograms generated using
only.
The structural differences between the 3 hy-
drates alluded to earlier were confirmed by X-ray

Fig. 9. Typical thermogram (DSC) produced by heating mag-

nesium stearate anhydrate (form B) at a rate of 2”C/min.
Fig. 8. Typical thermogram (DSC) produced by heating mag- The example shown is from a sample stored under 11.3%
nesium stearate trihydrate (form C) at a rate of 2O C/mm. relative humidity.
 177

form A

25
4
t5

2e
I i

Fig. 11. Typical X-ray diffraction pattern obtained from an

IO 5 ordered mount. The example shown is from a sample of
magnesium stearate trihydrate (form C).

ima were converted to crystal long spacings using

the Bragg equation:

nX=2d sin@ 0)

where d is the crystal long spacing, X is the

wavelength of the X-rays, 0 is the Bragg angle,
form B and n is the diffraction order (e.g. d/2, d/3,
d/4,. . . , ). The long spacing calculated for the
dihydrate is identical to that reported previously
(Miiller, 1977b, c); however, the long spacings
25 20 IS to 5 calculated for the anhydrate and trihydrate forms
are significantly greater than the values reported
28
previously for these hydration states (Table 4).
These data suggest that the crystal structures of
the anhydrate and trihydrate forms of magnesium
stearate isolated in the present work are different

TABLE 4

Crystal long spacings (d,,,) of the various forms of magnesium

stearate

Sample d,r (nm>

25 20 15 10 5
Form A (via ammonium soap) 5.220 f 0.003 (n = 20)
28 Form A (via sodium soap) 5.219 f 0.003 (n = 20)
Fig. 10. X-ray diffraction patterns obtained from random Form A (2.4 hydrate) 5.239 f 0.004 (n = 20)
powder mounts of the various forms of magnesium stearate. Form B 5.032 f 0.005 (n = 10)
Form C (52.8% R.H.) 5.270+0.003 (n =16)
Form C (80.9% R.H.) 5.304+0.003 (?a =18)
ordered powder mounts, since this type of mount Anhydrate * 3.55
minimized the contributions of the other lattice Dihydrate * 5.23
spacings and made the diffraction patterns easier Trihydrate * 5.13

to interpret. A typical diffraction pattern obtained Values are mean f S.E.M., n denotes the number of determina-
from an ordered powder mount is shown in Fig. tions.
11. Diffraction angles corresponding to peak max- * Values reported by Miiller (1977a).
178

TABLE 5

Summary of infrared absorption data for stearic acid and the various forms of magnesium stearate.

Wavenumber (cm-‘) Steak Acid Form A Form B Form C

3 600-3 100 (water of crystallization) _ s (w) s
3 300-2 500 (-OH stretch) _ _
;nbroad)
2 900 (C-H symmetric and m
asymmetric stretch) ;doublet) ;nhroad) ;doublet)
1700 (C=O stretch) s _ _ _
1616-l 540 (COO- asymmetric stretch) _ s S S

720 (CH, rocking)

;Ydoublet) (Wdoublet) zinglet) zinglet)

w, weak; m, moderate; s, strong.

from the crystal structures of corresponding correspond to monoclinic packing as well (Mtiller,
hydration states isolated in earlier studies. 1977b). Based on the appearance of this absorp-
The difference between the anhydrate and tri- tion band, the hydrocarbon chains in the stearic
hydrate forms isolated in this work and in earlier acid crystals possessed either a monoclinic or an
studies was further demonstrated by examining orthorhombic packing, which is consistent with
the infrared (IR) absorption spectra of the com- the packing determined for stearic acid by X-ray
pounds. The pertinent absorption data for stearic diffraction (von Sydow, 1956). The hydrocarbon
acid and the 3 hydrates of magnesium stearate chains in the dihydrate form of magnesium stearate
isolated in this work are summarized in Table 5. also existed in either a monoclinic or an ortho-
Because formation of the magnesium salt of stearic rhombic packing, which is again consistent with
acid occurred at the carbonyl end of the molecule, the packing previously described for this species
the observed differences between the absorption (Mtiller, 1977~). The triclinic or hexagonal packing
spectra of stearic acid and magnesium stearate are determined for the trihydrate, however, is not
mainly due to changes in the absorption bands consistent with the orthorhombic packing previ-
associated with the carbonyl group. For example, ously ascribed to this hydrate. The anhydrate has
spectra of all forms of magnesium stearate ex- previously been reported to possess hexagonal
hibited a shift from 1700 cm-’ to 1615-1540 packing (Vold, 1949; Mtiller, 1977c, 1982) which
cm-’ in the absorption band due to the carbonyl at first appeared to be consistent with the hexago-
stretch, indicating that the carbonyl group had nal or triclinic packing determined for the
gone from an unionized to an ionized state in anhydrate in this work. The angle of inclination of
forming the magnesium salt of stearic acid the hydrocarbon chains calculated from the X-ray
(Lecomte, 1950; Mehrotra, 1983). There were, data, however, was not consistent with hexagonal
however, several changes in the spectra associated packing, indicating that the hydrocarbon chains
with the aliphatic portion of the molecule. Of existed in a triclinic packing.
these, the absorption band appearing at approxi-
mately 720 cm-’ was of special interest. This
particular band is due to the CH, rocking mode, Conclusion
and its appearance (i.e. as a singlet or a doublet) is
related to the packing of the hydrocarbon chains Based on what has been discussed above, the
within the crystal (Chapman, 1957). Bands ap- following conclusions were drawn:
pearing as singlets correspond to either &clinic or 1. The crystal habit of magnesium stearate was
hexagonal packing, while bands appearing as not sensitive to either the pH of the precipitation
doublets correspond to orthorhombic packing, al- medium or the cooling rate, as all samples pre-
though it has been argued that a doublet may pared consisted of plate-shaped crystals. The
 179

needle-shaped crystals that have been reported References

previously could not be isolated under any condi-
tions. Billany, M.R. and Richards, J.H., Batch variation of mag-
2. The pH of the precipitation medium did nesium stearate and its effect on the dissolution rate of
salicylic acid from solid dosage forms. Drug Dee. Ind.
have an effect on the amount of moisture that was
Pharm., 8 (1982) 497-511.
bound within the magnesium stearate crystals. Callahan, J.C., Cleary, G.W., Kaplan, G., Kensler, T. and
Samples precipitated from media at pH 9 con- Nash, R.A., Equilibrium moisture content of pharmaceuti-
tained bound moisture equivalent to that expected cal excipients. Drug Deo. Ind. Pharm., 8 (1982) 355-369.
for a dihydrate, while samples precipitated from Chapman, D., The 720 cm-’ band in the infrared spectra of
crystalline long-chain compounds. J. Chem. Sot., (1957)
media at higher or lower pH values contained
448994491.
significantly less bound moisture. Colombo, I. and Carli, F., Comparative evaluation of structure
3. Drying the dihydrate at 105’C resulted in and micrometitics properties of magnesium stearate. II
the production of an anhydrous form of mag- Farmaco-Ed. Pr., 39 (1984) 329-341.
nesium stearate. The anhydrate retained the ap- De Boer, A.H., Bolhuis, G.K. and Lerk, C.F., Bonding char-
acteristics by scanning electron microscopy of powders
pearance of the original dihydrate; however, heat-
mixed with magnesium stearate. Powder Technol., 20 (1978)
ing resulted in a disruption of the crystals’ 3-di- 75-82.
mensional lattice structure. Handbook of Pharmaceutical Excipients, American Pharma-
4. The dihydrate form of magnesium stearate ceutical Association, Washington, DC, 1983, p. 173.
exhibited no tendency to gain or lose moisture Hanssen, D., Ftihrer, C. and Schafer, B., Beurteilung von
Magnesiumsterat als Tablettengleitmittel durch elektro-
under any of the relative humidity conditions used
nische Druckmessung. Pharm. Ind., 32 (1970) 97-101.
in this work. The anhydrate form sorbed moisture Holzer, A.W., Batch-to-batch variations of commercial mag-
readily, and rehydrated to a trihydrate when the nesium stearates. Labo-Pharma, Probl. Techn., 32 (1984)
relative humidity exceeded 50%. The trihydrate, 28-36.
like the dihydrate, did not exhibit any tendency to Hiilzer, A.W., Johansson, M. and Lindberg, N.-O., Some ob-
servations concerning the loss on drying of magnesium
lose moisture when stored under any of the rela-
stearate NF. Pharmocopeial Forum, May June (1985)
tive humidity conditions used in this work. 379-380.
5. The dihydrate form of magnesium stearate Iranloye, T.A. and Parrott, E.L., Effects of compression force,
isolated in the present work appears to be identi- particle size, and lubricants on dissolution rate. J. Pharm.
cal to previously reported dihydrates. This is not Sci., 67 (1978) 535-539.
Lecomte, J., Infra-red absorption spectra of metallic acetyl-
true for the anhydrate and &hydrate, as both of
acetonates. Disc. Faraday Sot., 9 (1950) 125-131.
these forms exhibit physical properties different Lerk, C.F. Bolhuis, G.K. and Smedema, S.S., Interaction of
from those described previously for corresponding lubricants and colloidal silica during mixing with excipi-
hydration states. ents. Pharm. Acta Helu., 52 (1977a) 33-39.
Lerk, C.F. and Bolhuis, G.K., Interaction of lubricants and
colloidal silica during mixing with excipients. Pharm. Acfa
Helu., 52 (1977b) 39-44.
Levy, G. and Gumtow, R.H., Effect of certain formulation
Acknowledgement
factors on dissolution rate of the active ingredient III.
Tablet Lubricants. J. Pharm. Sci., 52 (1963) 1139-1144.
The authors wish to thank Sandoz Research Mehrotra, K.N., Kachhwaha, R. and Singh, M., Studies on
Institute and Lederle Labs for their financial sup- beryllium soaps. Infrared absorption spectra and thermo-
gravimetric analysis. Thermochim. Acta, 62 (1983) 179-185.
port, and Merrell-Dow Research Institute for its
Miller, T.A., York, P. and Jones, T.M., Manufacture and
financial support and for the use of the thermal characterization of magnesium stearate and palmitate
analysis system. Special thanks go to Dr. SW. powders of high purity. J. Pharm. PharmacoL, 34 (1982)
Bailey for his help with the X-ray powder diffrac- 8P.
tion studies, and to Dr. J. Motzi and Dr. L. Miller, T.A. and York, P., Physical and chemical characteris-
tics of some high purity magnesium stearate and palmitate
Shenouda for their help in running the thermal
powders. I&. J. Pharm., 23 (1985) 55-67.
analysis system and for their comments and Miiller, B.W., Tribologische Gesetzmlssigkeiten und Erkent-
suggestions regarding interpretation of the data. nisse in der Tablettentechnologie. 3. Mitteilung: Un-
180

tersuchungen an reinen Magnesium- und Calciumstearaten. Shah, A.C. and Mlodozeniec, A.R., Mechanism of surface
Pharm. Ind., 39 (1977a) 161-165. lubrication: influence of lubricant-excipient mixing on
Miiller, B.W., The pseudo-polymorphism of magnesium processing characteristics of powders and properties of
stearate. Zbl. Pharm., 116 (1977b) 1261-1266. compressed tablets. J. Pharm. Sci., 66 (1977) 1377-1382.
Miiller, B.W., Die Pseudopolymorphie von Magnesiumsalzen Steffens, K.-J., Mliller, B.W. and List, P.H., Tribologische
hljherer Fettsauren. Arch. Pharm., 310 (1977~) 693-704. Gesetzmassigkeiten und Erkentnisse in der Tabletten-
Miiller, B.W., Steffens, K.-J., and List, P.H., Tribologische technologie. 7. Mitteilung: Untersuchungen an Mag-
Gesetzmlssigkeiten und Erkentnisse in der Tabletten- nesiumstearat-Handelspraparaten. Pharm. Ind., 44 (1982)
technologie. 6. Mitteilung; Einfllisse physikalischer Eigen- 826-830.
schaften chemisch reiner Magnesiumstearate auf die Strickland Jr., W.A., Nelson, E., Busse, L.W. and Higuchi, T.,
Schmeiereigenschaften bei der Tablettierung. Pharm. Ind., Physics of tablet compression IX. Fundamental aspects of
44 (1982) 729-734. tablet lubrication. J. Am. Pharm. Assoc. Sci. Ed., 45 (1956)
Nyqvist, H., Saturated salt solutions for maintaining specified 51-55.
relative humidities. Inf. J. Pharm. Tech&. Prod. Manuf, 4 Vold R.D. and Hattiangdi, G.S., Characterization of heavy
(1983) 47-48. metal soaps by X-ray diffraction. Ind. Eng. Chem., 41
Ragnarsson, G., HSlzer, A.W. and Sjiigren, J., The influence of (1949) 2311-2320.
mixing time and colloidal silica on the lubricating proper- Von Sydow, E., The normal fatty acids in solid state - a
ties of magnesium stearate. Int. J. Pharm., 3 (1979) 127-131. crystal structure investigation. Ark. Kern., 9 (1956) 231-254.
Rubinstein, M.H. and Moody, G., “Migration of Magnesium Weast, R.C. (Ed), Handbook of Chemistty and Physics, 58th
Stearate During Tablet Production”, 3rd International Con- edn., CRC, Cleveland, OH, 1977, p. E-46.
ference on Pharmaceutical Technology, Paris, 1983.