- Email: [email protected]
Metallographic Phenomena Observed in Amalgams
M E T A LL O G R A P H IC PH EN O M EN A OBSERVED IN A M A L G A M S . By Arthur W . Gray, Ph. D., Milford, Del.
(T r a n sa c tio n s o f the A m e rica n Institute o f M in ing E n gin eers, M ilw aukee M eeting, O ctober, 1918.)
P
HENOMENA which reveal Chemi I. CRUSHING STRENGTH. cal and physical processes that 1. T h e B lack D yn a m o m e t e r . probably go on in many alloys at Pract:call' all strength tes+s of dental '"mp-'ratures nof far removed from their m:lting ranges are observable in dental nirpk>ram,; hav° h-’r^ofor"1 been made amalgams. ^hese amalgams should, w:th the sprng d; nannnrt’r dev’sed by therefore, be of even more interest to the Dr. Black.4 The sp"c:m“ns tested were metallographer, the metallurgist, the in the form of cubes 0.085 in.(2.16 mm.) chcm’st, and the physicist than to the on eac^ edge. Thev were prepared by dentist who uses them for saving decay packing the amalgam in^o spl:t steel ing teeth. For this reason I am here de molds ess°nt:allv as it would be packed scribing a few of the results recently obtained during the course of a svste- 'n^o a footh cavitv, sorrK-'rrrs with s-'mpl" hand fnrc° with ma+;c investigaron of dental alio- s.1 In order to keep this account within Mows from a mallet. The cubes were reasonable bounds. I shall confine my crushed at whatever temperature the self almost entirely to the consideration room hnpp^ned to o^er. and the rrran of of strength in compression (crushing s“v°ral tr'als was taken to represent the strength) and of dimensional changes s^ne^h (Figures 1 and 2). during hardening (reaction expansion). Ev°rv physic’st w‘ll recognize that, Conditions influencing the mercurv con while such tests mav be of grpat value tent2 will receive some attention. A few in pioneer work, they are hardlv likelv results have already been communicated to vield results hav;ng the accuracy de to the American Physical Society.3 sirable in modern research. The cube, ’ In this in v estiga tion T w as assisted b y P a ris T. with its sharp corners and its radial C a rlisle. 4th, and b y M iss M artha C. B o r to n . To b oth nf these I am greatly indebted f o r the care asymmetry, is not the best shape for a w ith w h ich th ey look ed a fte r the n um erous e x p e r i m ental details a n d record s o f som e thousand s o f test specimen. The particular cubes te«ts the routin e m easurem ents f o r w h ich w ere used with Black’s dynamometer are so T'jide as a rule, b y y ou n g w om en w h o w ere train ed «specially f o r the duties a ssign ed to them . small that the accidental errors arsing 2T h is term is h ere used to m ean the p e r cent, o f from unavoidable irregularities in shape, m ercu ry in an am algam . See p. 517. 8A . W . G ray an d P . T. C a r lis le : T h e Influen ce size, and internal structure may easily o f T em perature u p on the C ru sh in g S trength o f a D ental A m algam . R och este r m eeting. P h ys. Rev. render the results uncertain. Their V o l. I I . P . 1 9 4 ; 1918. T h e In flu en ce o f the P r e s smallness introduces an error owing to sure and the T im e E m p loyed in C on densing a D e n tal A m algam u p on its C ru sh in g Strength at T em the restraining action of the jaws be peratu res betw een 10° and 100 °C ., P ittsb u rg h m eet i n g ; un p u blish ed. The In flu en ce o f A m alg am ation tween which they are crushed. TemperV a ria b les upon the M ercu ry C ontent and the C ru sh in g Strength o f a D ental A m algam . N ew m eeting. P h ys. R ev. V o l. 11, P . 4 9 2 ; 1918.
Y o rk 4D en tal C osm os, V o l. 37, P . 406 f f ; 1895.
513
TH E JOURNAL OP TH E N ATION AL DEN TAL ASSOCIATION. F ig u re 1.
S p rin g dyn a m om eter devised b y D r. B la c k f o r testing c r u s h in g stren gth an d flow o f D ental F illin g M aterials.
Figure 2.
M old s f o r p re p a rin g 0 .085-in . c u b e s o f am algam tested in B la c k d y n a m om eter. B e lo w are show n re la tive sizes o f cu b e s a n d c y lin d e rs u sed f o r c ru s h in g tests.
GRAY.— M ETALLO G RAPH IC PHENOM ENA IN AM ALGAM S.
ature affects the stiffness of the spring used for measuring the crushing force; it also affects considerably the strength of a dental amalgam. The strength of an amalgam specimen is also largely dependent upon the pressure under which it is molded into form; the hand
515
the load with regularity, thus permitting the beam to be kept balanced to a nicety right up to the moment of failure; and by the addition of electric heaters for bringing the specimen under test to the desired temperature, which was deter mined by a suitably placed thermo-ele-
F ig u re 3.
The
9000-kg.
O lsen m ach in e
ada pted to testin g strength o f m a te ria ls at a n y
tem perature a n d p ro v id e d w ith p o w e r driv e f o r a p p ly in g testing lo a d w ith regu larity.
or mallet packing cannot control this pressure sufficiently well for accurate comparisons of different alloys.
ment. Additional thermo-elements en abled temperature gradients within the heated region surrounding the cylinder to be adjusted to negligible values. The 2. Standardized P rocedure tor electromotive forces were measured by a Cr u sh in g - strength T ests . potentiometer. I have, therefore, discarded the old The test specimens were in the form Black dynamometer in favor of a 9000kg. Olsen testing machine, which was of right circular cylinders 10.04 mm. designed for operation by hand (Figure in diameter and of sufficient height to 3). This I modified by the addition of eliminate errors caused by the restrain an electric motor drive which applied ing action of the crushing jaws (Figure
516
T H E JO URNAL OF TH E N ATION AL DENTAL ASSOCIATION.
4). These cylinders were prepared by molding the freshly mixed amalgam in a thick-walled steel tube, the interior of which was highly polished. A measured force, applied by the testing machine and maintained for a measured time, was used in packing the amalgam between an accurately fitting piston and the remova ble bottom of the mold. This packing squeezed out .the excess of mercury and condensed the plastic mass into a firm, smooth c) Under with parallel ends. The procedure secured both uniformity of cross-section and uniformity of packing conditions.
during the testing. To save time in reaching the general conclusions desired, no attempt at controlling the temperature of the specimen during crushing was made in these earlier tests; the tests were made at whatever temperature the room happened to offer. As a consequence, the accidental errors, measured by devia tions from average values, were often considerably larger than in the later tests, in which temperature and other influencing conditions were carefully controlled. Even the preliminary exper iments demonstrated the need of a thor-
F ig u re 4.
T est cy lin d e rs o f am algam prepared b y a m algam atin g 61 gm. a llo y . C y lin d ers A an d B from lo w -s ilv e r a llo y com press an d cra ck . C y lin ders C , D, E , an d F fr o m high silver a llo y b u rst e x p lo s iv e ly . C y lin d ers B and D, b oth p a ck e d u n d e r 400 k g ., show d ifferen ce in size o f am algam from sam e w eight o f a llo y . D is 25 p e r cen t, la rg e r and 75 per cen t, s tron g er than B . C, D, and E are packed un d er 141, 400, and 1131 kg. per cir. cm ., respectiv ely . N atural s i z e : 10.0 mm. in diam eter.
Many preliminary experiments were made to obtain a general idea of the ways in which different conditions may influence the crushing strength of a den tal amalgam. These experiments showed the effect of variations in such factors as the proportions of mercury and alloy used in mixing the amalgam, the time devoted to triturating the mix, the tem perature of trituration, the pressure un der which the amalgam is molded into test cylinders, the time that this packing pressure is maintained, the height of the test , cylinder, the time that elapses be tween the making and the crushing of the cylinder, the temperature at which it is stored during this interval, and the rate at which the crushing load is applied
oly standardized technic ;f results of real value were to be obtained. 3.
Influence p ie c e
of
H e ig h t
of
T est-
U pon A pparen t Stre n g th .
In short-column tests it is important to have the test specimen sufficiently high to eliminate the effect of friction between the crushing jaws and the speci men; otherwise the apparent strength will be greater than the true strength of the material. Therefore, one of the first things to be determined was the mini mum height allowable with amalgam cjlinders 1 cm. in diameter. About 9 mm. was found to be ample; reduction to 6 mm. would not introduce serious
G RAY.— M ETALLO G RAPH IC PHENOM ENA IN AM ALGAM S.
517
error (Figure 5). However, no cylin ders shorter than 10 mm. were ever used.
strength,7 especially when the packing pressure was low. Our earliest experi ments indicated that it would be safe to 4. I nfluence of P acking P ressure standardize upon a packing time of 8 and of P acking T i m e . min. in order to eliminate possible irreg The next points to be decided were ularities arising from this source. Later the influence of packing pressure and of tests under controlled temperature showed packing time upon both the mercury the advisability of reducing the time to content and the strength of a high-grade 4 min.; still later ones resulted in adopt amalgam.5 The pressures used in pack ing 2. min. as the standard. The ing were systematically varied from 25 slight increase of strength observable to 3200 kg. per cir. cm.;0 and the time when a high packing pressure was mainF igu re 5.
Test sp ecim en s th at are to o sh o rt m ake m aterial a p p e a r stron ger than it r e a lly is.
during which a packing pressure was maintained was varied from 1 to 16 minutes. The time of packing was found to ex ert but little influence upon the crushing •'The ra tio o f the m asses o f m e rcu ry a n d a llo y that are m ix ed tog eth er to fo rm a n am algam w ill be fre q u e n tly r eferred to as the “ m e r cu ry -a llo y r a t io ." I n the p rocess o f m old in g th is am algam into test specim en s, som e o f th e m ercu ry , a lo n g w ith a sm all am ou n t o f a llo y , is squeezed ou t b y the pressure exerted thru the p is ton o f the m old. The p e rce n t age o f m ercu ry in the fin ish ed test specim en is h ere designated as the "m e r c u r y co n te n t.’ * cF o r con ven ien ce, n e a r ly a ll pressu res are here expressed in term s o f k ilo g ra m s-w e ig h t p e r c ir c u la r centim eter. 1 kg. p e r cir. c m . = £ kg. p e r cm .2 = 18.11 lb. per in .2 B la c k an d h is fo llo w e r s e x press cru sh in g stren gth in term s o f p c u n d s-w e ig h t u n d er w h ich a 0 .085-in . cu b e fa ile d . S u ch a result m u st tie m u ltip lied b y 7.6 to redu ce to kilo g ra m sw eig h t per c ir c u la r centim eter.
tained for a considerable time seemed traceable to cold working. This view was strengthened by results obtained at higher temperatures. Dilatometric meas urements of reaction expansion (which were found to be far more delicate indi cators than results of crashing tests) showed but little change upon reducing the packing time to one-quarter of a minute. The packing pressure was found to exert a marked influence (Figure 6). Even the earliest tests at room tempera 7T h ru o u t this p ap er the term "c r u s h in g stren gth ” is used to design ate the p ressu re (a lm o st a lw a y s expressed in kg. p e r c ir. cm .) u n d er w h ich the test specim en fa ile d w hen su b je cte d to sim ple a x ia l com pression .
518
T H E JO U RN AL OP TH E N A TION AL D EN TA L ASSOCIATION.
ture showed that, other things being equal, the crushing strength 5 increases directly as the logarithm of the packing pressure P; so that, within the limits of
nature of the amalgam and the tempera ture of crushing. At the same time, the mercury content was found to decrease logarithmically with increasing packing
F ig u re 6.
Packing Pressure in kg per Or cm B o th cru sh in g stren g th an d m e r cu ry -c o n te n t o f an am algam v a r y as lo g a rith m o f p a c k in g pressure. L a rg e d ev ia tion s from lo w e r curve arise fro m c r u s h in g specim en s at v a rio u s tem peratures.
experimental error, it can be expressed by a formula of the type 5 = A -f- B log P, which is equivalent to P = eaS_b, in which e stands for the logarithmic base, while A, B, a, and b, are paramet ers depending upon such factors as the
pressure; so that it, too, could be repre sented by a similar formula. The curves shown in Figure 6 were calculated from such equations. It can be seen that nearly all the points repre senting the results of the individual tests
G R A Y — M ETALLO G RAPH IC PH ENOM ENA IN AM ALGAM S.
lie close to the curves. A few, however, deviate by more than appears to be the uncertainty of our crushing tests. As was then assumed, and now confirmed by later experiments, these irregulari ties can be readily explained by the fact that the tests were made at room tem perature, on different days during the summer, which resulted not only in dif ferent cylinders being crushed at widely different temperatures, but also in differ ent amounts of moisture being condensed from the air upon the filings of alloy as they were being triturated with mercury. In addition, some of the deviations may have been due to the use of hand power in applying the crushing load, and to those occasional irregularities that seem inseparable from strength tests, such as defective specimens and failure to ad just properly between the jaws of the testing machine.8 The alloys used in all the other tests described in the second part of this paper were likewise amalgamated by trituration in a mortar, unless the con trary is stated. In most cases, however, 6 gm. of alloy were used, and changes were made in such factors as mercuryalloy ratio, trituration time, and age of the amalgam when tested. The logarithmic law connecting strength with packing pressure now has the support of more than 4000 crushing tests of amalgams prepared under widely varying conditions. The only systematic deviations from it thus far observed were found with cylinders undergoing transi tion: either the mix had not completely hardened, or else it was softening from 8E a ch c y lin d e r u sed in the tests r e co rd e d in F ig ure 6 w as prep ared fro m 5 gm . o f a h ig h -g ra d e a llo y in the fo rm o f filings, ju s t a s fu rn ish e d to the dentist. T h is w as in co rp o ra te d w ith 1.40 tim es its m ass o f purified m ercu ry b y th o r o ly tritu ratin g in a glass m orta r f o r 1.5 m in . T h e resu ltin g am algam w as q u ic k ly r o lle d in to a b a ll, d ropped into the m old, a n d -p a ck e d im m ediately. T he c y lin d ers w ere cru sh ed 24 hr. a fte r m aking.
519
the transformation that occurs near 70° C. (see p. 525.) Even in these cases the deviations were only slight. 5.
or L o g a r i t h m i c L a w . A s soon as the law was discovered it was applied for checking and adjusting observations. Instead of following the usual engineering procedure of averaging the results from several tests of speci mens all subjected to the same treatment, cylinders for testing a given point were always prepared in groups of three, or of five, packed under pressures increas ing in geometrical progression. The pres sures actually used were from the series 100, 141, 200, 283, 400, 566, 800, 1131, and 1600 kg. per cir. cm., in which \/2 is the common ratio. Then, if the crush ing strengths found for all the cylinders of a given group were plotted as equidis tant ordinates, the points obtained should lie close to a straight line. This pro cedure yields just as good an average value as if all the cylinders of the group were packed under the same pressure. In addition, it yields the particular curve connecting strength with packing pres sure that holds for the conditions under which the test is made. Its justification lies in the fact that, with the exceptions noted in the preceding paragraph, the logarithmic law has always been found to represent the observations within the limits of experimental error. A p p lic a tio n
6. I n f l u e n c e of A m a l g a m a t io n V a r i ables.
The fact that both the strength and the mercury content change regularly with the packing pressure suggests at once a close connection between the for mer and the latter. Long ago, Black found that the more mercury he squeezed out during the packing of a given amal gam, the stronger was the filling lie obtained; but his methods of experiment
520
TH E JO U RN AL OF TH E N A TION AL D EN TA L ASSOCIATION.
ing were not sufficiently precise to bring out any quantitative relationships. That the suggested connection is not of the nature of cause and effect, but is rather
of varying the mercury-alloy ratio from 0.5 to 2.5, while the trituration time was maintained uniformly at 1.5 min. This varied the mix from a very stiff to a very
F igu re 7.
Merctiry-Alloy Ratio
Packed under 141 kg
o f vary in g rela tive am ounts o f m e rcu ry an d a llo y that are m ix ed tog eth er in m aking an am algam .
a coincidence resulting from a common cause, becomes apparent as we notice how strength and mercury content are affected by varying such factors as the mercury-alloy ratio and the trituration time. Figures 7, 8 and 9 exhibit the results
pasty one. The cylinders packed under the lowest pressure (141 kg. per cir. cm.) show the phenomenon most clearly, both mercury content and crushing strength increasing rapidly in the same general way to maximum values, which remain constant as the mercury-alloy
G R A Y — M ETALLO G RAPH IC PHENOM ENA IN AM ALGAM S.
ratio is still further increased. The illus trations show how the increases o f pack ing pressure progressively wipe out the effects and cause the maxima to be
521
10 show that, in general, more mercury is retained in a test cylinder by prolong ing the trituration; but it is interesting to note that when the packing pressure
F ig u re 8.
Mercury-Alloy Ratio
Packed under 400 kg
M od ifica tion o f F ig u re 7 p r o d u ce d b y in cre a sin g p a ck in g p ressu re fro m 141 to 400 kg. p e r c ir. cm .
reached sooner; also, that constant mer cury content is reached before constant strength. Figures 10 and 11 exhibit the results o f varying the trituration time from 1 to 8 min., while a constant mercury ratio o f 1.40 was used. T he curves of Figure
is low, less mercury seems to be left in the cylinder by increasing the trituration time from 1 to 2 min. Apparently this is because the shorter trituration leaves many o f the alloy granules so large that a low -packing pressure is insufficient to squeeze out the free mercury from the
TH E JO U RN AL OF TH E N A TION AL DENTAL ASSOCIATION.
522
spaces among the solid particles. T he curves also show that for a given tritu ration time the mercury content changes
ing the time beyond this brings about a very gradual falling off in strength on account of the incipient setting o f the
F ig u re 9.
C Q O U &
c
35
30
Ä c o o
0.5
1.0
Mercury-AHoy Ratio
1.5
2.0
2.5
Packed under 1131 kg
M od ifica tion o f F ig u re s 7 and 8 b y fu rth e r in cre a se o f p a ck in g pressu re to 1131 kg. p e r c ir. cm.
almost inversely as the logarithm of the packing pressure. Increasing the trituration time while the packing pressure is kept constant is accompanied by a progressive increase in strength (Figure 11) until the latter reaches a maximum when the trituration is maintained for about 6 min. Prolong-
amalgam during the mixing. The re sults also indicate that the logarithmic law connecting crushing strength and packing pressure is applicable for any given trituration time within the range investigated. T he reason why excessive amalgamation produces so little effect upon the mercury content would appear
GRAY.— M ETALLO G RAPH IC PHENOM ENA IN AM ALGAM S.
523
F igu re 10.
Trituration Time C h an ges in m e rcu ry con te n t p ro d u ce d b y v a ry in g tim e devoted to tritu ra tin g m ix o f a llo y and m ercu ry .
F igu re 11.
G ain in c ru sh in g strength p ro d u ce d b y in cre a sin g tritu ra tio n tim e.
524
TH E JOURNAL OF TH E N A TION AL D EN TA L ASSOCIATION.
to be simply that the metals entering into a dental alloy are only very slightly soluble in mercury.0
7. I n f l u e n c e o f T e m p e r a t u r e . Figure 12 represents the results of crushing 63 similar cylinders of a high-
Figure 12.
Temperature Centigrade E ffe c t o f tem peratu re o n c ru s h in g strength. T ra n sitio n re g ion sh o w n b y the rap id f a ll in strength betw een 70° a n d 80° is confirm ed b y F igu res 13 an d 14.
eM cB a in a n d J oy n er (Jnl. Chem . S o c. L on d on V o l. 99, P . 1 9 5 -2 0 8 ; 1 9 1 1 ; D ental C osm os, V o l. 54, P . 6 4 6 -6 4 8 ; 1912) a n a ly z e d m e rcu ry th a t h a d been sa tu ra ted b y tw o w eek s’ c o n ta c t in a therm ostat w ith fillin gs o f s ilv e r -tin a llo y s , ran gin g b y 10 per cen t, step s fro m o n e p u re m etal to the other. T h e y fo u n d that a t 2 5 °C th e m e rcu ry disso lv e d b u t 0.044 p er cent, silv er a n d 0.75 p e r cent, tin , an d that these satu ration v a lu e s w ere indepen d en t o f the c om p o sition o f the a llo y . A t 6 3 °C ., a b o u t 0.18 per cen t, a n d 2.5 p e r cen t., resp e ctiv e ly , w ere dissolved . E v en th e m ost th o ro am algam ation o f d ental office p r a c tic e is h a r d ly lik e ly to b rin g a b o u t com plete sa tu ration .
grade dental amalgam at temperatures fairly uniformly distributed over the en tire range from below 25°C. to over 95° C. Each cylinder was prepared by incorporating the same mass of alloy filings with 1.60 times this mass of mer cury. After triturating for 4 min., the resulting smooth plastic amalgam was packed under a load of 400 kg. main-
G R A Y — M ETALLO G RAPH IC PHENOM ENA IN AM ALGAM S.
taincd for 8 min. This produced a cyl inder 10.04 mm. in diameter by 11.5 mm. high, 40 per cent, of its mass being mercury. All the cylinders prepared in this way were immediately placed in an incubator kept at 37.5°C. (body tem perature), where they remained for sev eral days before crushing, thus insuring practical completion of the hardening process. The alloy used contained approximately 68 per cent, silver, 26 per
525
pared, and also the precision with which crushing tests can be made, provided proper precautions be taken. The curve indicates that with rising temperature the crushing strength of the particular amalgam represented decreases somewhat faster than linearly from 5300 kg. per cm.2 at 25° to 4050 at 45°, and 2550 at 65°. At 37.5° the strength was found to be 4550 kg. per cm.2 or nearly 65,000 lb. per square inch.
F igu re 13.
100“
80
60
40
T ra n sition region rev ealed b y m odified R o b e rts -A u s te n m ethod. T em peratures plotted as ord in a te s are those o f n e u tra l b o d y . A b s cis s a s are galvanom eter d eflections. D ia g o n a l cu rves sh ow rate o f h e a tin g n e u tra l b o d y , each h o r i zon ta l sp ace representin g 20 m in.
8. T r a n s i t i o n R e g i o n N e a r 70° C. cent, tin, 5 per cent, copper, and 1 per cent. zinc. One of the most interesting features of All the 63 individual determinations this curve is the sudden plunge that it made on three separate days are in takes soon after passing 70 °C. Black cluded in the chart. The abscissa of a point represents the temperature of a had observed that an amalgam, upon cylinder at the time it was being crushed; heating to the temperature of boiling the ordinate, the compressive load in water, suffered a noticeable permanent kilograms-weight sustained by the cylin change in both appearance and volume, der at the instant of failure, and, there which he surmised was caused by some fore, the crushing strength in kilograms- thing other than ordinary thermal-exweight per circular centimeter. The pansion.10 McBain and Joyner, employ closeness with which the points follow a ing a liquid dilatometer in the usual smooth curve shows the uniformity with way, located a sharp transition at 71.5° which a dental amalgam can be pre’ »D ental C osm os, V o l. 37, P . 6 6 1 ; 1895.
526
T H E JOURNAL OP TH E NATION AL D EN TA L ASSOCIATION.
C.11 I have confirmed the existence of proportion of mercury. The beginning this transition by means of two other and the end of the transformation are entirely independent methods of thermal not sharply defined, but are in the neigh borhood of 70° and 90°, respectively. analysis. Figure 13 reproduces typical curves On cooling from 100° to 50° no corre obtained by a modification of the Rob- sponding evolution of heat was observed, erts-Austen method of observing temper showing that the transition is not a re ature differences between the sample versible one. The absorption of heat, and a neutral body while both were however, reappears at about the same being heated simultaneously in the same temperature with each successive warm enclosure. It is evident that the transi ing of the same specimen; but the magni tion is accompanied by an appreciable tude of the effect progressively dimin absorption of heat.12 The heating curves ishes. F igu re 14.
20
T ra n sition
30
40
50
60
70
80°C.
region revealed b y dila tom etrie m ethod. C u rves rep resen t h e a t ings o f sam e specim en on two s u cce ssiv e days.
obtained with two cylinders packed for 8 min. under 141 and 1131 kg. per cir. cm., respectively, showed sharp peaks at essentially the same temperature, which was close to 77.3° C., corresponding to about the completion of the sudden drop in strength. The magnitude of the heat absorption was somewhat greater with the cylinder packed at the lower pressure which also contained a greater “ Ib id ., V o l. 54, P . 6 4 9 ; 1912. 12M cB a in an d J oy n er (Ib id ., V o l. 54, P . 647) state that "C om p reh en s iv e s eries o f c o o lin g cu rve s o f the va riou s am algam s w ere taken, b u t th e ch em ica l ch an ges in v olv ed are too slow f o r th is m ethod to be adva n tageou s. I t w as fo u n d th at a ll am algam s o f tin a n d silver, even those th at h ave lo n g h a r dened, un d ergo p a r tia l fu s io n a t 65° to 1 0 0 °, an d g ra d u a lly b ecom e m ore flu id as the tem perature h ra ised .”
Figure 14 reproduces a pair of curves obtained by a method that depends upon variations in linear thermal expansivity. The two curves show the effect of heating the same specimen on successive days. The modifications produced by the first heating and the reappearance of the sharp minimum should be noted. The amalgams represented by Figures 13 and 14 were prepared from different alloys 9.
In flu en ce at
of
P a c k in g
P ressu re
V a r io u s T e m p e r a t u r e s .
Having learned how the strength of a high-grade dental amalgam could be altered by varying separately either the packing pressure or the temperature of
G RAY.— M ETALLO G RAPH IC PH ENOM ENA IN AM ALGAM S.
crushing, it became of interest to investi gate the effect of simultaneously varying both of these parameters within wide limits. Accordingly, a group of test
527
Some of the results are represented by Figure IS. The curve for each tempera ture was calculated from an equation of the type S = A -J- B log P, in which
F igu re 15.
L og a rith m ic la w
c o n n e ctin g c ru sh in g stren gth a n d p a ck in g pressure betw een 2 6° a n d 65° C.
cylinders were packed for 4 min. under loads of 100, 200, 400, 800, and 1600 kg. per cir. cm. After storing at 37.5° for 3 days, these were crushed at temper atures distributed over the interval from 25° to 100° C.
h o ld s a t a n y tem perature
the letters have the same meaning as on p. 518. The plotted points, each of which represents an individual crush (not an average of several tests), were located after applying small corrections to re duce the observations to what they would
528
TH E JO U RN AL OF TH E N A TION AL D EN TA L ASSOCIATION.
have been if the temperatures at the time o f crushing had been exactly those stated along the right-hand margin. Precisely the same data are plotted
results of tests at 70° and at 75°, which are here added, indicate slight devia tions from the logarithmic law at these temperatures; but these deviations are
F igu re 16.
Packing Pressure in kg per cir cm T ra n sform a tion o f lo g a rith m ic cu rv e s o f F ig u re 15 b y p lo ttin g ab scissa s p r o p o r tio n a l to rithm s o f p a ck in g pressures.
in Figure 16. The only difference is that the curves o f Figure 15 have been transformed into straight lines by the simple mathematical expedient of making the abscissas proportional to the loga rithms o f the packing pressures. The
lo g a
hardly greater than the experimental errors. Still another view o f these same facts is afforded by Figure 17. Here again the straight lines (each of which corre sponds to a definite packing pressure,
G RAY.— M ETALLO G RAPH IC PH EN OM ENA IN AM ALGAM S.
ranging from 100 kg. per cir. cm. at the bottom to 1600 at the top) result from direct mathematical transformation of the lines appearing in the preceding two
529
sure being indicated by its own symbol. Additional observations near 85° and 100° are included. It is possible that the lines represent-
F ig u re 17.
D ata rep resen ted in F igu res 15 a n d 16 p lo tte d to show th a t strength o f g iv en a m algam specim en d ecrea ses lin e a rly w ith in cre a se o f tem perature. L o g a rith m ic law is expressed b y equal a n g u la r d iv erg en ce o f stra igh t lin es co rre sp o n d in g to v a rio u s p a ck in g pressures.
figures, the logarithmic relations being expressed by the equal angular spacing. In this plot no corrections are necessary for temperature deviations from the nom inal values. T he points represent the actual observations, each packing pres
ing the results between 25° and 65° in Figure 17 should be slightly curved as in Figure 12. However, the straight lines, as drawn, afford a simple means of expressing the observed facts within the limits o f experimental error. The
530
T H E JO U RN AL OF TH E N A TION AL D EN TA L ASSOCIATION.
reason why the first part o f the curve in Figure 12 lies above the median line o f Figure 17 is to be sought in the fact that the cylinders used for the former
changes with change of temperature. Perhaps this has some connection with the transition o f tin from the white to the gray form.
F igu re 18.
H ardening
o f am algam s.
A m algam
fro m h ig h -s ilv e r a llo y fro m lo w -s ilv e r a llo y .
is
75 p e r cen t,
stron ger
than
that
F igu re 19.
j25ooo cn «
Pr.sKurM
£4000
iflk g p t r c ir c c m
'5 i.
• > 400
*•3000
a* 141
£2000
Io> looo 7
503
A g e of Amalgam in Days
7
488
A m algam s fro m den tal a llo y o f g o o d q u a lity sh ow n o re trog ression in strength 'd u rin g m ore than 1% year.
were somewhat harder, because they were subjected to the packing pressure for 8 min. instead of 4, and they were stored more than twice as long before crushing. Cooling the amalgam below 20° or 25° was found to result in a marked reduction o f the rate at which strength
10.
Influence and
of
A ge
upon
A lloy
A m algam .
T w o more questions o f vital interest to the dentist w ill now be considered. (1 ) H ow does the strength vary with the time that elapses after the plastic m ix is condensed into a rigid cylinder by the application o f the packing-pressure ?
G R A Y — M ETALLO G RAPH IC PH EN OM ENA IN AM ALGAM S.
Does an amalgam, after hardening to its maximum strength, maintain this strength indefinitely? (2) Does the alloy from which the dentist prepares his amalgam deteriorate from long standing on the dealer’s shelves? In Figure 18, the upper curve shows the behavior of an amalgam from the same high-grade alloy that has been represented in the preceding charts; the lower curve shows the behavior of an amalgam from a typical low-grade alloy still much used by dentists all over
531
months before crushing. Whether the mix was triturated for 1.5 or for 6 min., whether it was packed under 100, 400, or 1600 kg., there is not the slightest indication of retrogression in strength. The results demonstrate conclusively that it is perfectly possible to make a dental alloy which will yield a stable amalgam. Figure 20 shows the uniformity of product attainable in the regular com mercial manufacture of a high-grade dental alloy. It shows, also, that the
F igu re 20.
-,5ooo «4 0 0 0
1600 400
;3 0 0 0
? 200 0
1912 1913 1914 D ate o f Manufacture
1915
1916
1917
1918
U n iform ity an d s ta b ility o f d en tal a llo y o f g o o d qu a lity. F ig u res to rig h t o f lin es in d ica te pressu res u n d e r w h ich am algam s w ere m o ld e d in to test c y lin ders.
the country. Both alloys were treated in exactly the same way; the cylinders had been packed for 2 min. under 400 kg. per cir. cm. The results with other packing pressures yielded similar curves, which are here omitted to avoid confus ing the chart. For the same reason, those cylinders between 0.1 hr. and 6 hr. old are also omitted. Note the extremely rapid rise in strength during the first 2 hr., followed by a slow increase that continues for more than a week. Note, also, the striking difference between the high-grade and the low-grade alloy. The stability of the high-grade amal gam is well brought out by Figure 19, which compares cylinders only 7 days old with others which had been stored at body temperature for more than 16
sample of alloy which had been stand ing in the package as long as 6 years yielded just as strong an amalgam as the sample which had been made only a few days before testing. These results prove beyond question that a properly made alloy does not deteriorate with lapse of time, provided it has been kept reasonably clean and has not been sub jected to gross abuse such as storing in an unusually warm place.13 13W a rd con d u cte d an exten sive series o f tests w h ic h le d him to the c o n c lu s io n th at som e, at least, o f th e h ig h -s ilv e r d en tal a llo y s on the m a r ket show m arked de te rio ra tio n w ith time b o th b e fo re a n d a fte r b e in g m ade into am algam fillings. D en tal R ev iew , V o l. 21, P . 4 1 6 ; 1907. D ental C o s m os, V o l. 50, P . 1 1 4 -1 1 6 ; 1908. Jnl. A llie d D ental S o c ., V o l. 11, P . 446 1916.
(T o be continued in a later issue.)
(T r a n sa c tio n s o f the A m e rica n Institute o f M in ing E n gin eers, M ilw aukee M eeting, O ctober, 1918.)
P
HENOMENA which reveal Chemi I. CRUSHING STRENGTH. cal and physical processes that 1. T h e B lack D yn a m o m e t e r . probably go on in many alloys at Pract:call' all strength tes+s of dental '"mp-'ratures nof far removed from their m:lting ranges are observable in dental nirpk>ram,; hav° h-’r^ofor"1 been made amalgams. ^hese amalgams should, w:th the sprng d; nannnrt’r dev’sed by therefore, be of even more interest to the Dr. Black.4 The sp"c:m“ns tested were metallographer, the metallurgist, the in the form of cubes 0.085 in.(2.16 mm.) chcm’st, and the physicist than to the on eac^ edge. Thev were prepared by dentist who uses them for saving decay packing the amalgam in^o spl:t steel ing teeth. For this reason I am here de molds ess°nt:allv as it would be packed scribing a few of the results recently obtained during the course of a svste- 'n^o a footh cavitv, sorrK-'rrrs with s-'mpl" hand fnrc° with ma+;c investigaron of dental alio- s.1 In order to keep this account within Mows from a mallet. The cubes were reasonable bounds. I shall confine my crushed at whatever temperature the self almost entirely to the consideration room hnpp^ned to o^er. and the rrran of of strength in compression (crushing s“v°ral tr'als was taken to represent the strength) and of dimensional changes s^ne^h (Figures 1 and 2). during hardening (reaction expansion). Ev°rv physic’st w‘ll recognize that, Conditions influencing the mercurv con while such tests mav be of grpat value tent2 will receive some attention. A few in pioneer work, they are hardlv likelv results have already been communicated to vield results hav;ng the accuracy de to the American Physical Society.3 sirable in modern research. The cube, ’ In this in v estiga tion T w as assisted b y P a ris T. with its sharp corners and its radial C a rlisle. 4th, and b y M iss M artha C. B o r to n . To b oth nf these I am greatly indebted f o r the care asymmetry, is not the best shape for a w ith w h ich th ey look ed a fte r the n um erous e x p e r i m ental details a n d record s o f som e thousand s o f test specimen. The particular cubes te«ts the routin e m easurem ents f o r w h ich w ere used with Black’s dynamometer are so T'jide as a rule, b y y ou n g w om en w h o w ere train ed «specially f o r the duties a ssign ed to them . small that the accidental errors arsing 2T h is term is h ere used to m ean the p e r cent, o f from unavoidable irregularities in shape, m ercu ry in an am algam . See p. 517. 8A . W . G ray an d P . T. C a r lis le : T h e Influen ce size, and internal structure may easily o f T em perature u p on the C ru sh in g S trength o f a D ental A m algam . R och este r m eeting. P h ys. Rev. render the results uncertain. Their V o l. I I . P . 1 9 4 ; 1918. T h e In flu en ce o f the P r e s smallness introduces an error owing to sure and the T im e E m p loyed in C on densing a D e n tal A m algam u p on its C ru sh in g Strength at T em the restraining action of the jaws be peratu res betw een 10° and 100 °C ., P ittsb u rg h m eet i n g ; un p u blish ed. The In flu en ce o f A m alg am ation tween which they are crushed. TemperV a ria b les upon the M ercu ry C ontent and the C ru sh in g Strength o f a D ental A m algam . N ew m eeting. P h ys. R ev. V o l. 11, P . 4 9 2 ; 1918.
Y o rk 4D en tal C osm os, V o l. 37, P . 406 f f ; 1895.
513
TH E JOURNAL OP TH E N ATION AL DEN TAL ASSOCIATION. F ig u re 1.
S p rin g dyn a m om eter devised b y D r. B la c k f o r testing c r u s h in g stren gth an d flow o f D ental F illin g M aterials.
Figure 2.
M old s f o r p re p a rin g 0 .085-in . c u b e s o f am algam tested in B la c k d y n a m om eter. B e lo w are show n re la tive sizes o f cu b e s a n d c y lin d e rs u sed f o r c ru s h in g tests.
GRAY.— M ETALLO G RAPH IC PHENOM ENA IN AM ALGAM S.
ature affects the stiffness of the spring used for measuring the crushing force; it also affects considerably the strength of a dental amalgam. The strength of an amalgam specimen is also largely dependent upon the pressure under which it is molded into form; the hand
515
the load with regularity, thus permitting the beam to be kept balanced to a nicety right up to the moment of failure; and by the addition of electric heaters for bringing the specimen under test to the desired temperature, which was deter mined by a suitably placed thermo-ele-
F ig u re 3.
The
9000-kg.
O lsen m ach in e
ada pted to testin g strength o f m a te ria ls at a n y
tem perature a n d p ro v id e d w ith p o w e r driv e f o r a p p ly in g testing lo a d w ith regu larity.
or mallet packing cannot control this pressure sufficiently well for accurate comparisons of different alloys.
ment. Additional thermo-elements en abled temperature gradients within the heated region surrounding the cylinder to be adjusted to negligible values. The 2. Standardized P rocedure tor electromotive forces were measured by a Cr u sh in g - strength T ests . potentiometer. I have, therefore, discarded the old The test specimens were in the form Black dynamometer in favor of a 9000kg. Olsen testing machine, which was of right circular cylinders 10.04 mm. designed for operation by hand (Figure in diameter and of sufficient height to 3). This I modified by the addition of eliminate errors caused by the restrain an electric motor drive which applied ing action of the crushing jaws (Figure
516
T H E JO URNAL OF TH E N ATION AL DENTAL ASSOCIATION.
4). These cylinders were prepared by molding the freshly mixed amalgam in a thick-walled steel tube, the interior of which was highly polished. A measured force, applied by the testing machine and maintained for a measured time, was used in packing the amalgam between an accurately fitting piston and the remova ble bottom of the mold. This packing squeezed out .the excess of mercury and condensed the plastic mass into a firm, smooth c) Under with parallel ends. The procedure secured both uniformity of cross-section and uniformity of packing conditions.
during the testing. To save time in reaching the general conclusions desired, no attempt at controlling the temperature of the specimen during crushing was made in these earlier tests; the tests were made at whatever temperature the room happened to offer. As a consequence, the accidental errors, measured by devia tions from average values, were often considerably larger than in the later tests, in which temperature and other influencing conditions were carefully controlled. Even the preliminary exper iments demonstrated the need of a thor-
F ig u re 4.
T est cy lin d e rs o f am algam prepared b y a m algam atin g 61 gm. a llo y . C y lin d ers A an d B from lo w -s ilv e r a llo y com press an d cra ck . C y lin ders C , D, E , an d F fr o m high silver a llo y b u rst e x p lo s iv e ly . C y lin d ers B and D, b oth p a ck e d u n d e r 400 k g ., show d ifferen ce in size o f am algam from sam e w eight o f a llo y . D is 25 p e r cen t, la rg e r and 75 per cen t, s tron g er than B . C, D, and E are packed un d er 141, 400, and 1131 kg. per cir. cm ., respectiv ely . N atural s i z e : 10.0 mm. in diam eter.
Many preliminary experiments were made to obtain a general idea of the ways in which different conditions may influence the crushing strength of a den tal amalgam. These experiments showed the effect of variations in such factors as the proportions of mercury and alloy used in mixing the amalgam, the time devoted to triturating the mix, the tem perature of trituration, the pressure un der which the amalgam is molded into test cylinders, the time that this packing pressure is maintained, the height of the test , cylinder, the time that elapses be tween the making and the crushing of the cylinder, the temperature at which it is stored during this interval, and the rate at which the crushing load is applied
oly standardized technic ;f results of real value were to be obtained. 3.
Influence p ie c e
of
H e ig h t
of
T est-
U pon A pparen t Stre n g th .
In short-column tests it is important to have the test specimen sufficiently high to eliminate the effect of friction between the crushing jaws and the speci men; otherwise the apparent strength will be greater than the true strength of the material. Therefore, one of the first things to be determined was the mini mum height allowable with amalgam cjlinders 1 cm. in diameter. About 9 mm. was found to be ample; reduction to 6 mm. would not introduce serious
G RAY.— M ETALLO G RAPH IC PHENOM ENA IN AM ALGAM S.
517
error (Figure 5). However, no cylin ders shorter than 10 mm. were ever used.
strength,7 especially when the packing pressure was low. Our earliest experi ments indicated that it would be safe to 4. I nfluence of P acking P ressure standardize upon a packing time of 8 and of P acking T i m e . min. in order to eliminate possible irreg The next points to be decided were ularities arising from this source. Later the influence of packing pressure and of tests under controlled temperature showed packing time upon both the mercury the advisability of reducing the time to content and the strength of a high-grade 4 min.; still later ones resulted in adopt amalgam.5 The pressures used in pack ing 2. min. as the standard. The ing were systematically varied from 25 slight increase of strength observable to 3200 kg. per cir. cm.;0 and the time when a high packing pressure was mainF igu re 5.
Test sp ecim en s th at are to o sh o rt m ake m aterial a p p e a r stron ger than it r e a lly is.
during which a packing pressure was maintained was varied from 1 to 16 minutes. The time of packing was found to ex ert but little influence upon the crushing •'The ra tio o f the m asses o f m e rcu ry a n d a llo y that are m ix ed tog eth er to fo rm a n am algam w ill be fre q u e n tly r eferred to as the “ m e r cu ry -a llo y r a t io ." I n the p rocess o f m old in g th is am algam into test specim en s, som e o f th e m ercu ry , a lo n g w ith a sm all am ou n t o f a llo y , is squeezed ou t b y the pressure exerted thru the p is ton o f the m old. The p e rce n t age o f m ercu ry in the fin ish ed test specim en is h ere designated as the "m e r c u r y co n te n t.’ * cF o r con ven ien ce, n e a r ly a ll pressu res are here expressed in term s o f k ilo g ra m s-w e ig h t p e r c ir c u la r centim eter. 1 kg. p e r cir. c m . = £ kg. p e r cm .2 = 18.11 lb. per in .2 B la c k an d h is fo llo w e r s e x press cru sh in g stren gth in term s o f p c u n d s-w e ig h t u n d er w h ich a 0 .085-in . cu b e fa ile d . S u ch a result m u st tie m u ltip lied b y 7.6 to redu ce to kilo g ra m sw eig h t per c ir c u la r centim eter.
tained for a considerable time seemed traceable to cold working. This view was strengthened by results obtained at higher temperatures. Dilatometric meas urements of reaction expansion (which were found to be far more delicate indi cators than results of crashing tests) showed but little change upon reducing the packing time to one-quarter of a minute. The packing pressure was found to exert a marked influence (Figure 6). Even the earliest tests at room tempera 7T h ru o u t this p ap er the term "c r u s h in g stren gth ” is used to design ate the p ressu re (a lm o st a lw a y s expressed in kg. p e r c ir. cm .) u n d er w h ich the test specim en fa ile d w hen su b je cte d to sim ple a x ia l com pression .
518
T H E JO U RN AL OP TH E N A TION AL D EN TA L ASSOCIATION.
ture showed that, other things being equal, the crushing strength 5 increases directly as the logarithm of the packing pressure P; so that, within the limits of
nature of the amalgam and the tempera ture of crushing. At the same time, the mercury content was found to decrease logarithmically with increasing packing
F ig u re 6.
Packing Pressure in kg per Or cm B o th cru sh in g stren g th an d m e r cu ry -c o n te n t o f an am algam v a r y as lo g a rith m o f p a c k in g pressure. L a rg e d ev ia tion s from lo w e r curve arise fro m c r u s h in g specim en s at v a rio u s tem peratures.
experimental error, it can be expressed by a formula of the type 5 = A -f- B log P, which is equivalent to P = eaS_b, in which e stands for the logarithmic base, while A, B, a, and b, are paramet ers depending upon such factors as the
pressure; so that it, too, could be repre sented by a similar formula. The curves shown in Figure 6 were calculated from such equations. It can be seen that nearly all the points repre senting the results of the individual tests
G R A Y — M ETALLO G RAPH IC PH ENOM ENA IN AM ALGAM S.
lie close to the curves. A few, however, deviate by more than appears to be the uncertainty of our crushing tests. As was then assumed, and now confirmed by later experiments, these irregulari ties can be readily explained by the fact that the tests were made at room tem perature, on different days during the summer, which resulted not only in dif ferent cylinders being crushed at widely different temperatures, but also in differ ent amounts of moisture being condensed from the air upon the filings of alloy as they were being triturated with mercury. In addition, some of the deviations may have been due to the use of hand power in applying the crushing load, and to those occasional irregularities that seem inseparable from strength tests, such as defective specimens and failure to ad just properly between the jaws of the testing machine.8 The alloys used in all the other tests described in the second part of this paper were likewise amalgamated by trituration in a mortar, unless the con trary is stated. In most cases, however, 6 gm. of alloy were used, and changes were made in such factors as mercuryalloy ratio, trituration time, and age of the amalgam when tested. The logarithmic law connecting strength with packing pressure now has the support of more than 4000 crushing tests of amalgams prepared under widely varying conditions. The only systematic deviations from it thus far observed were found with cylinders undergoing transi tion: either the mix had not completely hardened, or else it was softening from 8E a ch c y lin d e r u sed in the tests r e co rd e d in F ig ure 6 w as prep ared fro m 5 gm . o f a h ig h -g ra d e a llo y in the fo rm o f filings, ju s t a s fu rn ish e d to the dentist. T h is w as in co rp o ra te d w ith 1.40 tim es its m ass o f purified m ercu ry b y th o r o ly tritu ratin g in a glass m orta r f o r 1.5 m in . T h e resu ltin g am algam w as q u ic k ly r o lle d in to a b a ll, d ropped into the m old, a n d -p a ck e d im m ediately. T he c y lin d ers w ere cru sh ed 24 hr. a fte r m aking.
519
the transformation that occurs near 70° C. (see p. 525.) Even in these cases the deviations were only slight. 5.
or L o g a r i t h m i c L a w . A s soon as the law was discovered it was applied for checking and adjusting observations. Instead of following the usual engineering procedure of averaging the results from several tests of speci mens all subjected to the same treatment, cylinders for testing a given point were always prepared in groups of three, or of five, packed under pressures increas ing in geometrical progression. The pres sures actually used were from the series 100, 141, 200, 283, 400, 566, 800, 1131, and 1600 kg. per cir. cm., in which \/2 is the common ratio. Then, if the crush ing strengths found for all the cylinders of a given group were plotted as equidis tant ordinates, the points obtained should lie close to a straight line. This pro cedure yields just as good an average value as if all the cylinders of the group were packed under the same pressure. In addition, it yields the particular curve connecting strength with packing pres sure that holds for the conditions under which the test is made. Its justification lies in the fact that, with the exceptions noted in the preceding paragraph, the logarithmic law has always been found to represent the observations within the limits of experimental error. A p p lic a tio n
6. I n f l u e n c e of A m a l g a m a t io n V a r i ables.
The fact that both the strength and the mercury content change regularly with the packing pressure suggests at once a close connection between the for mer and the latter. Long ago, Black found that the more mercury he squeezed out during the packing of a given amal gam, the stronger was the filling lie obtained; but his methods of experiment
520
TH E JO U RN AL OF TH E N A TION AL D EN TA L ASSOCIATION.
ing were not sufficiently precise to bring out any quantitative relationships. That the suggested connection is not of the nature of cause and effect, but is rather
of varying the mercury-alloy ratio from 0.5 to 2.5, while the trituration time was maintained uniformly at 1.5 min. This varied the mix from a very stiff to a very
F igu re 7.
Merctiry-Alloy Ratio
Packed under 141 kg
o f vary in g rela tive am ounts o f m e rcu ry an d a llo y that are m ix ed tog eth er in m aking an am algam .
a coincidence resulting from a common cause, becomes apparent as we notice how strength and mercury content are affected by varying such factors as the mercury-alloy ratio and the trituration time. Figures 7, 8 and 9 exhibit the results
pasty one. The cylinders packed under the lowest pressure (141 kg. per cir. cm.) show the phenomenon most clearly, both mercury content and crushing strength increasing rapidly in the same general way to maximum values, which remain constant as the mercury-alloy
G R A Y — M ETALLO G RAPH IC PHENOM ENA IN AM ALGAM S.
ratio is still further increased. The illus trations show how the increases o f pack ing pressure progressively wipe out the effects and cause the maxima to be
521
10 show that, in general, more mercury is retained in a test cylinder by prolong ing the trituration; but it is interesting to note that when the packing pressure
F ig u re 8.
Mercury-Alloy Ratio
Packed under 400 kg
M od ifica tion o f F ig u re 7 p r o d u ce d b y in cre a sin g p a ck in g p ressu re fro m 141 to 400 kg. p e r c ir. cm .
reached sooner; also, that constant mer cury content is reached before constant strength. Figures 10 and 11 exhibit the results o f varying the trituration time from 1 to 8 min., while a constant mercury ratio o f 1.40 was used. T he curves of Figure
is low, less mercury seems to be left in the cylinder by increasing the trituration time from 1 to 2 min. Apparently this is because the shorter trituration leaves many o f the alloy granules so large that a low -packing pressure is insufficient to squeeze out the free mercury from the
TH E JO U RN AL OF TH E N A TION AL DENTAL ASSOCIATION.
522
spaces among the solid particles. T he curves also show that for a given tritu ration time the mercury content changes
ing the time beyond this brings about a very gradual falling off in strength on account of the incipient setting o f the
F ig u re 9.
C Q O U &
c
35
30
Ä c o o
0.5
1.0
Mercury-AHoy Ratio
1.5
2.0
2.5
Packed under 1131 kg
M od ifica tion o f F ig u re s 7 and 8 b y fu rth e r in cre a se o f p a ck in g pressu re to 1131 kg. p e r c ir. cm.
almost inversely as the logarithm of the packing pressure. Increasing the trituration time while the packing pressure is kept constant is accompanied by a progressive increase in strength (Figure 11) until the latter reaches a maximum when the trituration is maintained for about 6 min. Prolong-
amalgam during the mixing. The re sults also indicate that the logarithmic law connecting crushing strength and packing pressure is applicable for any given trituration time within the range investigated. T he reason why excessive amalgamation produces so little effect upon the mercury content would appear
GRAY.— M ETALLO G RAPH IC PHENOM ENA IN AM ALGAM S.
523
F igu re 10.
Trituration Time C h an ges in m e rcu ry con te n t p ro d u ce d b y v a ry in g tim e devoted to tritu ra tin g m ix o f a llo y and m ercu ry .
F igu re 11.
G ain in c ru sh in g strength p ro d u ce d b y in cre a sin g tritu ra tio n tim e.
524
TH E JOURNAL OF TH E N A TION AL D EN TA L ASSOCIATION.
to be simply that the metals entering into a dental alloy are only very slightly soluble in mercury.0
7. I n f l u e n c e o f T e m p e r a t u r e . Figure 12 represents the results of crushing 63 similar cylinders of a high-
Figure 12.
Temperature Centigrade E ffe c t o f tem peratu re o n c ru s h in g strength. T ra n sitio n re g ion sh o w n b y the rap id f a ll in strength betw een 70° a n d 80° is confirm ed b y F igu res 13 an d 14.
eM cB a in a n d J oy n er (Jnl. Chem . S o c. L on d on V o l. 99, P . 1 9 5 -2 0 8 ; 1 9 1 1 ; D ental C osm os, V o l. 54, P . 6 4 6 -6 4 8 ; 1912) a n a ly z e d m e rcu ry th a t h a d been sa tu ra ted b y tw o w eek s’ c o n ta c t in a therm ostat w ith fillin gs o f s ilv e r -tin a llo y s , ran gin g b y 10 per cen t, step s fro m o n e p u re m etal to the other. T h e y fo u n d that a t 2 5 °C th e m e rcu ry disso lv e d b u t 0.044 p er cent, silv er a n d 0.75 p e r cent, tin , an d that these satu ration v a lu e s w ere indepen d en t o f the c om p o sition o f the a llo y . A t 6 3 °C ., a b o u t 0.18 per cen t, a n d 2.5 p e r cen t., resp e ctiv e ly , w ere dissolved . E v en th e m ost th o ro am algam ation o f d ental office p r a c tic e is h a r d ly lik e ly to b rin g a b o u t com plete sa tu ration .
grade dental amalgam at temperatures fairly uniformly distributed over the en tire range from below 25°C. to over 95° C. Each cylinder was prepared by incorporating the same mass of alloy filings with 1.60 times this mass of mer cury. After triturating for 4 min., the resulting smooth plastic amalgam was packed under a load of 400 kg. main-
G R A Y — M ETALLO G RAPH IC PHENOM ENA IN AM ALGAM S.
taincd for 8 min. This produced a cyl inder 10.04 mm. in diameter by 11.5 mm. high, 40 per cent, of its mass being mercury. All the cylinders prepared in this way were immediately placed in an incubator kept at 37.5°C. (body tem perature), where they remained for sev eral days before crushing, thus insuring practical completion of the hardening process. The alloy used contained approximately 68 per cent, silver, 26 per
525
pared, and also the precision with which crushing tests can be made, provided proper precautions be taken. The curve indicates that with rising temperature the crushing strength of the particular amalgam represented decreases somewhat faster than linearly from 5300 kg. per cm.2 at 25° to 4050 at 45°, and 2550 at 65°. At 37.5° the strength was found to be 4550 kg. per cm.2 or nearly 65,000 lb. per square inch.
F igu re 13.
100“
80
60
40
T ra n sition region rev ealed b y m odified R o b e rts -A u s te n m ethod. T em peratures plotted as ord in a te s are those o f n e u tra l b o d y . A b s cis s a s are galvanom eter d eflections. D ia g o n a l cu rves sh ow rate o f h e a tin g n e u tra l b o d y , each h o r i zon ta l sp ace representin g 20 m in.
8. T r a n s i t i o n R e g i o n N e a r 70° C. cent, tin, 5 per cent, copper, and 1 per cent. zinc. One of the most interesting features of All the 63 individual determinations this curve is the sudden plunge that it made on three separate days are in takes soon after passing 70 °C. Black cluded in the chart. The abscissa of a point represents the temperature of a had observed that an amalgam, upon cylinder at the time it was being crushed; heating to the temperature of boiling the ordinate, the compressive load in water, suffered a noticeable permanent kilograms-weight sustained by the cylin change in both appearance and volume, der at the instant of failure, and, there which he surmised was caused by some fore, the crushing strength in kilograms- thing other than ordinary thermal-exweight per circular centimeter. The pansion.10 McBain and Joyner, employ closeness with which the points follow a ing a liquid dilatometer in the usual smooth curve shows the uniformity with way, located a sharp transition at 71.5° which a dental amalgam can be pre’ »D ental C osm os, V o l. 37, P . 6 6 1 ; 1895.
526
T H E JOURNAL OP TH E NATION AL D EN TA L ASSOCIATION.
C.11 I have confirmed the existence of proportion of mercury. The beginning this transition by means of two other and the end of the transformation are entirely independent methods of thermal not sharply defined, but are in the neigh borhood of 70° and 90°, respectively. analysis. Figure 13 reproduces typical curves On cooling from 100° to 50° no corre obtained by a modification of the Rob- sponding evolution of heat was observed, erts-Austen method of observing temper showing that the transition is not a re ature differences between the sample versible one. The absorption of heat, and a neutral body while both were however, reappears at about the same being heated simultaneously in the same temperature with each successive warm enclosure. It is evident that the transi ing of the same specimen; but the magni tion is accompanied by an appreciable tude of the effect progressively dimin absorption of heat.12 The heating curves ishes. F igu re 14.
20
T ra n sition
30
40
50
60
70
80°C.
region revealed b y dila tom etrie m ethod. C u rves rep resen t h e a t ings o f sam e specim en on two s u cce ssiv e days.
obtained with two cylinders packed for 8 min. under 141 and 1131 kg. per cir. cm., respectively, showed sharp peaks at essentially the same temperature, which was close to 77.3° C., corresponding to about the completion of the sudden drop in strength. The magnitude of the heat absorption was somewhat greater with the cylinder packed at the lower pressure which also contained a greater “ Ib id ., V o l. 54, P . 6 4 9 ; 1912. 12M cB a in an d J oy n er (Ib id ., V o l. 54, P . 647) state that "C om p reh en s iv e s eries o f c o o lin g cu rve s o f the va riou s am algam s w ere taken, b u t th e ch em ica l ch an ges in v olv ed are too slow f o r th is m ethod to be adva n tageou s. I t w as fo u n d th at a ll am algam s o f tin a n d silver, even those th at h ave lo n g h a r dened, un d ergo p a r tia l fu s io n a t 65° to 1 0 0 °, an d g ra d u a lly b ecom e m ore flu id as the tem perature h ra ised .”
Figure 14 reproduces a pair of curves obtained by a method that depends upon variations in linear thermal expansivity. The two curves show the effect of heating the same specimen on successive days. The modifications produced by the first heating and the reappearance of the sharp minimum should be noted. The amalgams represented by Figures 13 and 14 were prepared from different alloys 9.
In flu en ce at
of
P a c k in g
P ressu re
V a r io u s T e m p e r a t u r e s .
Having learned how the strength of a high-grade dental amalgam could be altered by varying separately either the packing pressure or the temperature of
G RAY.— M ETALLO G RAPH IC PH ENOM ENA IN AM ALGAM S.
crushing, it became of interest to investi gate the effect of simultaneously varying both of these parameters within wide limits. Accordingly, a group of test
527
Some of the results are represented by Figure IS. The curve for each tempera ture was calculated from an equation of the type S = A -J- B log P, in which
F igu re 15.
L og a rith m ic la w
c o n n e ctin g c ru sh in g stren gth a n d p a ck in g pressure betw een 2 6° a n d 65° C.
cylinders were packed for 4 min. under loads of 100, 200, 400, 800, and 1600 kg. per cir. cm. After storing at 37.5° for 3 days, these were crushed at temper atures distributed over the interval from 25° to 100° C.
h o ld s a t a n y tem perature
the letters have the same meaning as on p. 518. The plotted points, each of which represents an individual crush (not an average of several tests), were located after applying small corrections to re duce the observations to what they would
528
TH E JO U RN AL OF TH E N A TION AL D EN TA L ASSOCIATION.
have been if the temperatures at the time o f crushing had been exactly those stated along the right-hand margin. Precisely the same data are plotted
results of tests at 70° and at 75°, which are here added, indicate slight devia tions from the logarithmic law at these temperatures; but these deviations are
F igu re 16.
Packing Pressure in kg per cir cm T ra n sform a tion o f lo g a rith m ic cu rv e s o f F ig u re 15 b y p lo ttin g ab scissa s p r o p o r tio n a l to rithm s o f p a ck in g pressures.
in Figure 16. The only difference is that the curves o f Figure 15 have been transformed into straight lines by the simple mathematical expedient of making the abscissas proportional to the loga rithms o f the packing pressures. The
lo g a
hardly greater than the experimental errors. Still another view o f these same facts is afforded by Figure 17. Here again the straight lines (each of which corre sponds to a definite packing pressure,
G RAY.— M ETALLO G RAPH IC PH EN OM ENA IN AM ALGAM S.
ranging from 100 kg. per cir. cm. at the bottom to 1600 at the top) result from direct mathematical transformation of the lines appearing in the preceding two
529
sure being indicated by its own symbol. Additional observations near 85° and 100° are included. It is possible that the lines represent-
F ig u re 17.
D ata rep resen ted in F igu res 15 a n d 16 p lo tte d to show th a t strength o f g iv en a m algam specim en d ecrea ses lin e a rly w ith in cre a se o f tem perature. L o g a rith m ic law is expressed b y equal a n g u la r d iv erg en ce o f stra igh t lin es co rre sp o n d in g to v a rio u s p a ck in g pressures.
figures, the logarithmic relations being expressed by the equal angular spacing. In this plot no corrections are necessary for temperature deviations from the nom inal values. T he points represent the actual observations, each packing pres
ing the results between 25° and 65° in Figure 17 should be slightly curved as in Figure 12. However, the straight lines, as drawn, afford a simple means of expressing the observed facts within the limits o f experimental error. The
530
T H E JO U RN AL OF TH E N A TION AL D EN TA L ASSOCIATION.
reason why the first part o f the curve in Figure 12 lies above the median line o f Figure 17 is to be sought in the fact that the cylinders used for the former
changes with change of temperature. Perhaps this has some connection with the transition o f tin from the white to the gray form.
F igu re 18.
H ardening
o f am algam s.
A m algam
fro m h ig h -s ilv e r a llo y fro m lo w -s ilv e r a llo y .
is
75 p e r cen t,
stron ger
than
that
F igu re 19.
j25ooo cn «
Pr.sKurM
£4000
iflk g p t r c ir c c m
'5 i.
• > 400
*•3000
a* 141
£2000
Io> looo 7
503
A g e of Amalgam in Days
7
488
A m algam s fro m den tal a llo y o f g o o d q u a lity sh ow n o re trog ression in strength 'd u rin g m ore than 1% year.
were somewhat harder, because they were subjected to the packing pressure for 8 min. instead of 4, and they were stored more than twice as long before crushing. Cooling the amalgam below 20° or 25° was found to result in a marked reduction o f the rate at which strength
10.
Influence and
of
A ge
upon
A lloy
A m algam .
T w o more questions o f vital interest to the dentist w ill now be considered. (1 ) H ow does the strength vary with the time that elapses after the plastic m ix is condensed into a rigid cylinder by the application o f the packing-pressure ?
G R A Y — M ETALLO G RAPH IC PH EN OM ENA IN AM ALGAM S.
Does an amalgam, after hardening to its maximum strength, maintain this strength indefinitely? (2) Does the alloy from which the dentist prepares his amalgam deteriorate from long standing on the dealer’s shelves? In Figure 18, the upper curve shows the behavior of an amalgam from the same high-grade alloy that has been represented in the preceding charts; the lower curve shows the behavior of an amalgam from a typical low-grade alloy still much used by dentists all over
531
months before crushing. Whether the mix was triturated for 1.5 or for 6 min., whether it was packed under 100, 400, or 1600 kg., there is not the slightest indication of retrogression in strength. The results demonstrate conclusively that it is perfectly possible to make a dental alloy which will yield a stable amalgam. Figure 20 shows the uniformity of product attainable in the regular com mercial manufacture of a high-grade dental alloy. It shows, also, that the
F igu re 20.
-,5ooo «4 0 0 0
1600 400
;3 0 0 0
? 200 0
1912 1913 1914 D ate o f Manufacture
1915
1916
1917
1918
U n iform ity an d s ta b ility o f d en tal a llo y o f g o o d qu a lity. F ig u res to rig h t o f lin es in d ica te pressu res u n d e r w h ich am algam s w ere m o ld e d in to test c y lin ders.
the country. Both alloys were treated in exactly the same way; the cylinders had been packed for 2 min. under 400 kg. per cir. cm. The results with other packing pressures yielded similar curves, which are here omitted to avoid confus ing the chart. For the same reason, those cylinders between 0.1 hr. and 6 hr. old are also omitted. Note the extremely rapid rise in strength during the first 2 hr., followed by a slow increase that continues for more than a week. Note, also, the striking difference between the high-grade and the low-grade alloy. The stability of the high-grade amal gam is well brought out by Figure 19, which compares cylinders only 7 days old with others which had been stored at body temperature for more than 16
sample of alloy which had been stand ing in the package as long as 6 years yielded just as strong an amalgam as the sample which had been made only a few days before testing. These results prove beyond question that a properly made alloy does not deteriorate with lapse of time, provided it has been kept reasonably clean and has not been sub jected to gross abuse such as storing in an unusually warm place.13 13W a rd con d u cte d an exten sive series o f tests w h ic h le d him to the c o n c lu s io n th at som e, at least, o f th e h ig h -s ilv e r d en tal a llo y s on the m a r ket show m arked de te rio ra tio n w ith time b o th b e fo re a n d a fte r b e in g m ade into am algam fillings. D en tal R ev iew , V o l. 21, P . 4 1 6 ; 1907. D ental C o s m os, V o l. 50, P . 1 1 4 -1 1 6 ; 1908. Jnl. A llie d D ental S o c ., V o l. 11, P . 446 1916.
(T o be continued in a later issue.)