s 
 
 14. GS: 
 CIR &i£- 
 
 c -3 
 
 j£je~£ \ .^*f 
 
 3 
 
 STATE OF ILLINOIS 
 
 DEPARTMENT OF REGISTRATION AND EDUCATION 
 
 THERMAL EXPANSION 
 OF CERTAIN ILLINOIS 
 LIMESTONES AND DOLOMITES 
 
 Richard D. Harvey 
 
 ILLINOIS GEOLOGIC* 
 
 8Ui<VE v ' ' p ne ' 
 A PR 211986 
 
 ILLINOIS STATE GEOLOGICAL SURVEY 
 John C. Frye, Chief URBANA 
 
 CIRCULAR 415 1967 
 
THERMAL EXPANSION OF CERTAIN 
 ILLINOIS LIMESTONES AND DOLOMITES 
 
 Richard D. Harvey 
 
 ABSTRACT 
 
 Thermal expansion is an important property of lime- 
 stones that affects their utilization as building stone and 
 possibly as aggregates in concrete. Thermal expansion 
 curves were obtained on 82 essentially pure calcitic speci- 
 mens taken from 21 different Illinois limestones and on 62 Il- 
 linois dolomites and dolomitic limestone specimens taken 
 from 18 different samples. The specimens were 1 inch in di- 
 ameter and 4 inches long. Their expansion was measured 
 dry, with a vitreous silica dilatometer, over a temperature 
 range of -4 to 17 6° F. The average of the mean expansion 
 coefficient of limestones in the interval -4 to 68° F. is 1.9x 
 10- 6 /°F.; in the interval 68 to 176° F. it is 4.5 x lO" 6 /" F. 
 The corresponding coefficients of the dolomites are 4.3 x 
 10" 6 / F. and 5.5 x 10" 6 /° F. 
 
 The data indicate that the texture of the rock affects 
 the thermal expansion behavior of limestones and dolomites 
 that has not heretofore been noted. Most striking is an ob- 
 servation that coarse-grained limestones and, to some ex- 
 tent, coarse-grained dolomites expand at a significantly low- 
 er rate below 40° F. than above 80° F. Fine-grained lime- 
 stones and dolomites, however, expand at nearly the same 
 rate throughout the natural climatic temperature range in- 
 vestigated. Expansion measurements taken in the direction 
 perpendicular and parallel to the bedding show: (1) all do- 
 lomite samples expanded most perpendicular to the bedding, 
 (2) all coarse-grained crinoidal limestone samples expand- 
 ed most perpendicular to the bedding, and (3) many fine- 
 grained limestone samples expanded most parallel to the 
 bedding. 
 
 Increased rock expansion was also observed as the 
 dolomite, quartz, and clay mineral content of the rock in- 
 creases. 
 
2 ILLINOIS STATE GEOLOGICAL SURVEY CIRCULAR 415 
 
 INTRODUCTION 
 
 Limestones and dolomites, as well as other materials, expand and con- 
 tract when exposed to climatic temperature changes. The importance of thermal 
 expansion has long been recognized in the design of large concrete structures and 
 buildings, particularly as it affects stresses and joints (Mitchell, 1956). Ther- 
 mal expansion of aggregates is also a factor in the durability of concrete (Pear- 
 son, 1942; Hornebrook, 1942; and Callan, 1950). 
 
 Because of the importance of thermal expansion in the use of carbonate 
 rocks as aggregates and as dimension stone, this study was initiated to deter- 
 mine the expansion characteristics of selected limestones and dolomites from Il- 
 linois deposits. Variations in mineral composition and texture are common in 
 these rocks, and such variations are independent and potentially important dura- 
 bility factors. This report covers the results of tests on dry, mostly pure calcitic 
 limestones and relatively pure dolomites that have a wide variety of textures. 
 The temperature range over which the expansion measurements were taken was 
 -4 to 176° F. in order to include the climatic conditions of winter and summer 
 seasons in Illinois. 
 
 Acknowledgments 
 
 The author wishes to acknowledge the assistance of Mr. R. K. Kirby of 
 the U. So National Bureau of Standards, Thermal Expansion Laboratory, in obtain- 
 ing a metal alloy specimen for which the thermal expansion had been previously 
 determined in the Bureau's laboratory. Professor Albert V. Carozzi, of the Uni- 
 versity of Illinois, helpfully commented on the manuscript. 
 
 Previous Investigations 
 
 Early studies of the thermal expansion of carbonate rocks included tests 
 of marbles by Bartlett (1832), Adie (1836), Hallock (1891), Nisi (1913), Wheeler 
 (1910), and Souder and Hidnert (1919). Souder and Hidnert also tested specimens 
 of Indiana Bedford (Salem) Limestone. Griffith (1936) measured thermal expansion 
 of many rock types in the temperature range 32 to 500° F. He observed that the 
 mean coefficient of linear expansion of limestones in the range from room temp- 
 erature to 212° F. varied from 1. 8 to 6. 1 x lO" 6 /" F. The rate of heating of Grif- 
 fith's specimens was 1.7 to 2.5° F. per minute. 
 
 Willis and DeReuse (193 9) measured the expansion of several kinds of 
 rock used as concrete aggregates in wet and dry conditions between 37 and 140° 
 F. and noted that some limestones expanded linearly with increased temperature 
 and others expanded in a distinctly nonlinear manner. Johnson and Parsons (1944) 
 determined the mean expansion coefficient of 13 relatively pure limestones and 
 found it to vary from 0.5 to 5. 1 x 10"^ with an average of 2.5 x 10 _ 6/° F. between 
 -4 and 140° F. Their rate of heating was 0.9° F. per minute. The average of the 
 mean expansion coefficient of the impure limestones was 4. 6 x 10" . The average 
 of the mean coefficient of six dolomites tested, including four from Illinois (table 1), 
 was 4.3 x 10"^. Mather et al. (1953) reported the expansion of several limestones 
 used for concrete aggregates. They used the strain gage method of Callan (195 0) 
 to measure the total expansion from 35 to 135° F. Their results of samples from 
 
THERMAL EXPANSION OF LIMESTONES AND DOLOMITES 3 
 
 Illinois deposits are listed in table 1, Selected portions of the data included 
 herein were previously given in other reports (Harvey, 1966; Harvey, in press). 
 
 Several studies of the thermal expansion of concrete aggregates, concrete, 
 and cement have been made to determine the influence of expansion on the dura- 
 bility of concrete. Volume changes in hardened Portland cement were measured 
 by Davis (1930) and later by Meyers (1940, 1951). Their results indicate that 
 the expansion of cement varies considerably, depending on its composition, wa- 
 ter content, and age. From his expansion data on cement, three limestones, and 
 concrete, Koenitzer (1936) concluded that one cannot predict the expansion of the 
 concrete from measurements on the individual components. 
 
 Callan (1950) suspected that the deterioration of concrete during freeze - 
 thaw tests was influenced by the thermal properties of the aggregates and meas- 
 ured the expansion of several aggregate types as well as mortars used as cement- 
 ing matrix. He found that as the difference in mean expansion coefficient between 
 the aggregate and mortar increased, the freeze-thaw durability decreased. The 
 results of Walker, Bloem, and Mullen (1952) do not confirm this correlation; how- 
 ever, they conclude that the expansion of concrete in many instances can be pre- 
 dicted by knowledge of the expansion of the components in the concrete. These 
 contrasting results probably arise from significant differences in the samples 
 
 TABLE 1 - THERMAL EXPANSION DATA ON ILLINOIS LIMESTONES 1 ' 2 AND 
 DOLOMITES 3 REPORTED IN PREVIOUS LITERATURE 
 
 Description 
 
 Location 
 
 Expansion coefficient (per °F.) 
 
 In bedding 
 
 Across bedding 
 
 L bedding 
 
 Average 
 
 Limestone, dense, 
 
 
 mottled 
 
 Krause 
 
 Limestone, dense 
 
 Krause 
 
 Limestone, cherty 
 
 Krause 
 
 Limestone, argil- 
 
 
 laceous with 
 
 
 laminar struc- 
 
 
 ture 
 
 Krause 
 
 2.1 x 10 
 
 2.4 
 
 2.0 
 
 5.4 
 
 -6 
 
 1.9 x 1 
 
 2.3 
 1.9 
 
 5.2 
 
 -6 
 
 1.8 x 10 -6 1.9 x 10 -6 
 
 1.8 
 1.6 
 
 5.0 
 
 2.2 
 
 1.8 
 
 5.2 
 
 Limestone, argil- 
 laceous with 
 globular struc- 
 ture 
 
 Limestone 
 
 Krause 
 Alton 
 
 3.7 
 
 4.1 
 
 2.7 
 
 3.5 
 
 2.4 
 
 Dolomite gravel 
 (four speci- 
 mens) 
 
 Elgin 
 
 4.4 
 
 ^From St. Louis Limestone. 
 
 2 Data of Mather et al., 1953. Temp, range 35 to 135° F. 
 
 3 Data of Johnson and Parsons, 1944. Temp, range -4 to 140° F. 
 
1 ILLINOIS STATE GEOLOGICAL SURVEY CIRCULAR 415 
 
 studied. Walker, Bloem, and Mullen also conclude that the concretes having 
 higher expansion coefficients were less resistant to temperature changes than 
 concretes with lower coefficients. 
 
 Dunn (1963) gives results of thermal expansion of limestones used as ag- 
 gregates in New York. His measurements were made between -90 and 160° F. 
 using a recording dilatometer described by Rosenholtz and Smith (194 9). Dunn 
 found little or no correlation between the thermal expansion of the limestones 
 tested and the results of freeze-thaw tests; however, the expansion was found 
 to be related to the mineral content of the rock. A good review of the literature 
 and discussion of the significance of thermal expansion of concrete aggregates 
 is given by Cook (195 6). 
 
 FUNDAMENTAL ASPECTS OF THERMAL EXPANSION 
 
 The cohesion of solid crystals or the forces holding the atoms or ions to- 
 gether are generally considered in terms of bonding between atomic particles. 
 Where the atomic particles are ions, as in calcite crystals (CaCO^) in limestone, 
 this bonding is brought about by electrostatic attraction of negatively charged 
 CO3" 2 ions and positively charged Ca +2 ions. The C and O in the CO3 ions are 
 held together by covalent type bonds. 
 
 Albert Einstein showed that atoms and ions in solids actually vibrate with- 
 in crystals, and as the temperature is increased, the amplitude of these vibrations 
 increases with the frequency remaining constant. The increased vibration am- 
 plitude takes the form of strain waves moving through the crystal, distorting the 
 relative positions of the ions, and usually results in an increase in the volume 
 of the solid. These mobile crystal defects can be altered by impurities, grain 
 boundaries, and other defects that may be present. 
 
 In some crystals, increased temperature produces a reduction in length 
 along certain directions in the crystal; this phenomenon is termed negative ex- 
 pansion. According to Austin et al. (1940), calcite expands greatly along the 
 c-axis and contracts along the a-axes. Austin explains this behavior in the fol- 
 lowing way. The carbonate ions have a flat triangular shape in calcite and are 
 aligned in planes perpendicular to the c-axis at low temperatures. At higher tem- 
 peratures, the ions rotate out of perpendicularity, thereby stretching the c-axis 
 and allowing the equatorial axes to contract. 
 
 Mean Linear Expansion Coefficient (Expansivity) 
 
 The amount of expansion that a solid undergoes depends on its total length 
 and the number of degrees of temperature change. A useful value to know is the 
 coefficient of expansion, which is the amount of expansion of each unit length 
 for each degree change in temperature. For example, in figure 1 a bar is shown 
 whose length is L at temperature t , and whose length is Lt at some higher tem- 
 perature. The difference, Lt - L =AL, is the total amount of expansion. The 
 expansion of each unit of length, Al/L , is related to the difference in tempera- 
 ture, t - t = At, according to 
 
 4^=* At, (1) 
 
THERMAL EXPANSION OF LIMESTONES AND DOLOMITES 
 
 h- 
 
 AL 
 
 t Y///////////////1 
 
 where OC is the linear coefficient of ex- 
 pansion. A graph of experimental val- 
 ues of AL/L Q plotted against t for most 
 materials, as is diagramatically shown 
 in figure 2, has a curvature to the plot- 
 ted points, indicating that oC is not a 
 constant but is dependent upon particu- 
 lar values of t. A more precise equation 
 of expansion curves takes the form 
 
 AL 
 
 = at + bt^ + 
 
 (2) 
 
 Figure 1 - Linear expansion of a bar from 
 temperature t Q to t. 
 
 Rearrangement gives 
 
 Lt = 
 
 L (1 + at + bt^ + . . .) 
 
 (3) 
 
 The constants a and b can be determined from the experimental points of AL/L 
 and t (Hidnert and Souder, 1950), and the expansion occurring between any two 
 temperatures within the interval of validity of (2) can be determined by calculating 
 the unit expansions, AL/L Q , for both values of t and subtracting to find their dif- 
 ference: 
 
 In practice, the expansion is assumed to be linear (the dashed line in fig. 2), and 
 equation (1) is used. To indicate this simplification, OC is termed the mean ex- 
 pansion coefficient or expansivity; the latter will be used in this report, and it 
 is calculated from experimental data as follows: 
 
 OC = 
 
 (4) 
 
 - ti 
 
 This method of analysis will give accurate expansion predictions only for the tem- 
 perature range t-^ to t2» 
 
 SAMPLES 
 
 Test Specimens 
 
 Large blocks of relatively pure, freshly quarried limestones were collected 
 from 13 quarries, and similar blocks of dolomite and dolomitic limestones were 
 
ILLINOIS STATE GEOLOGICAL SURVEY CIRCULAR 415 
 
 (^1 
 
 AL 
 Lo 
 
 AL 
 
 collected from 12 quarries (table 2). 
 Three of the limestone quarries and 
 four dolomite quarries contained two 
 or more distinctly different types of 
 stone, and each type was sampled. 
 Two limestone samples were collected 
 from outcrops, and three limestone sam- 
 ples came from drill cores, making a 
 total of 21 limestone and 18 dolomite 
 samples tested. The samples selected 
 were free of joints and other obvious 
 imperfections. 
 
 With a few exceptions, four 
 specimens, 1 inch in diameter and 5 
 inches long, were cored from each sam- 
 ple block. Two cores were taken per- 
 pendicular and two parallel to the bed- 
 ding of the blocks. Only three test 
 
 specimens perpendicular to bedding were cored from the three subsurface cores. 
 The ends of the specimens were sawed with a diamond saw and ground smooth and 
 parallel, in a specially designed grinder, to a length of 4 inches. Most of the 
 specimens were allowed to dry naturally in the laboratory for a month or more; 
 others were placed under slight vacuum for one week and stored in a desiccator 
 for one week prior to testing. No differences in expansion resulted from the two 
 drying processes. 
 
 Figure 2 - Expansion curve: graph of unit 
 expansion versus temperature. 
 
 Mineral Composition and Grain Size 
 
 Pieces from the sample blocks were subjected to X-ray and HC1 residue 
 analyses. The results are listed in table 3 in order of decreasing purity. All of 
 the limestones consist of more than 90 percent calcite except sample 62-24, which 
 contains approximately 84 percent calcite. The dolomitic limestones are 65-37B, 
 65-37C, and 65-31, with 70, 55, and 60 percent calcite, respectively. All the 
 dolomites have 85 percent or more of the mineral dolomite. 
 
 The textural characteristics of each sample were observed in thin sections 
 cut perpendicular to the bedding, using a polarizing microscope. Examples of the 
 textures are shown in plates 1 through 4. All of the photomicrographs are shown 
 at the same magnification except figure 2, plate 3. The latter is more highly mag- 
 nified to illustrate in greater detail the texture of the very fine-grained limestones 
 and similar material present in varying amounts in most other limestone samples- 
 that is, the dark material in plates 1 and 2 and in figure 2, plate 3. 
 
 As the hardness of limestones is notably influenced by the particle size 
 of the individual crystalline grains (Harvey, 1963), it was suspected that the 
 grain size of the specimen also would affect its thermal expansion. To obtain a 
 measure of the grain size, the maximum dimensions of 100 grains or more were 
 grouped into logarithmic size categories, and the median grain size was deter- 
 mined for each sample from cumulative frequency diagrams. The grains measured 
 were randomly chosen by microscopic point-count procedures. Such methods are 
 subject to certain bias, according to Van der Plas (1962), yet the method does 
 give valid comparisons of the grain size of the samples. 
 
THERMAL EXPANSION OF LIMESTONES AND DOLOMITES 
 
 TABLE 2 - SOURCE AND DESCRIPTION OF SAMPLES 
 
 Sample 
 no. 
 
 Location 
 (nearest town) 
 
 Formation- 1 
 member 
 
 Description 
 
 Median 
 grain 
 size 2 (mm) 
 
 LIMESTONES 
 
 JN-12 Omega, Q~ 
 
 Omega 
 
 Fine— grained with scattered 
 sparry calcite fossil frag- 
 ments 
 
 0.010 
 
 65-1 Fairmount, Q 
 
 Livingston Pseudobreccia ; patchy network 0.010 
 of sparry calcite mosaics set 
 in very fine-grained calcite; 
 microfossiliferous 
 
 62-2 4M Buncombe, Q Kinkaid 
 
 62-24 Buncombe, Q Kinkaid 
 
 Fine grained, microfossilif- 
 erous, with scattered coarse- 
 grained crinoid fragments 
 
 0.020 
 
 Fine grained; microfossils 0.010 
 and fossil fragments occur in 
 thin laminations 
 
 62-22 Anna, Q 
 
 Ste. Genevieve Oolitic limestone; medium- to 
 Fredonia coarse-grained calcite sur- 
 rounded by very fine-grained 
 calcite, cemented by sparry 
 calcite 
 
 0.010 
 
 K-14 Alton, Q 
 
 63-39 Prairie du 
 Rocher, Q 
 
 63-40 Prairie du 
 Rocher , Q 
 
 62-25 Mill Creek, Q 
 
 62-26 Quincy, Q 
 
 65-3 2U Monmouth, Q 
 
 (upper level) 
 
 St. Louis Very fine-grained and equant 0.003 
 granular limestone 
 
 Salem Very fine-grained fossil 0.030 
 
 Rocher fragments cemented by 
 
 sparry calcite; few large 
 crinoid grains present 
 
 Salem Fine- and medium- 0.04 
 
 Kidd grained fossils cemented 
 with sparry calcite 
 
 Ullin Coarse grained; crinoidal, 0.06 
 
 Harrodsburg with secondary overgrowths 
 interlayered with abundant 
 fine crystalline bryozoan 
 fragments 
 
 Burlington Coarse-grained crinoid frag- 0.22 
 "Quincy bed" ments with secondary over- 
 growths and very fine-grained 
 bryozoan fragments 
 
 Burlington Coarse grained, crinoidal, 0.54 
 with recrystallized microspar 
 matrix 
 
8 ILLINOIS STATE GEOLOGICAL SURVEY CIRCULAR 415 
 
 TABLE 2 - Continued 
 
 Sample 
 no. 
 
 Location 
 (nearest town) 
 
 Forma tion-'- 
 member 
 
 Description 
 
 Median 
 
 grain 
 
 size^(mm) 
 
 LIMESTONES (cont.) 
 
 62-32L Monmouth, Q Burlington 
 (lower level) 
 
 65-33 Milan, Q 
 
 C2740 Phillipstown 
 W5464 4 
 
 RA-7 Baldwin 
 W1600 
 
 64-12 Gale, Q 
 
 63-42 Valmeyer, Q 
 
 C4774 Ashley 
 W4400 
 
 Wapsipinicon 
 Davenport 
 
 Lower 
 Devonian 
 
 Moccasin 
 Springs 
 "reef" 
 
 St. Clair 
 
 62-27 Thebes, Otc. 5 Girardeau 
 
 63-41 Valmeyer, Q Kimmswick 
 
 "buff" 
 
 Kimmswick 
 
 65-42 Thebes, Otc. Kimmswick 
 
 Kimmswick 
 
 Coarse-grained mosaic 
 with patches of fine- 
 grained bryozoan fragments 
 
 Very fine-grained and equant 
 granular limestone 
 
 Coarse-grained crinoidal, 
 unequant mosaic with 
 few laminations of 
 finer particles 
 
 Coarse-grained crystalline 
 mosaic of unequant calcite 
 particles 
 
 Very fine grained and equant 
 grains with few scattered 
 medium-grained fossil 
 fragments 
 
 Very fine grained and equant 
 grains with scattered fine 
 crystalline grains 
 
 Coarse grained; mosaic of 
 partly recrystallized crinoid 
 fragments with some fine- 
 grained bryozoan fragments and 
 microspar 
 
 Coarse grained; mosaic of 
 partly recrystallized crinoid 
 fragments and few finely 
 recrystallized bryozoans 
 
 Coarse-grained mosaic 
 of unequant particles, 
 and scattered patches 
 of fine-grained microspar 
 and fossil fragments 
 
 Coarse-grained, crinoidal, 
 with numerous recrystal- 
 lized bryozoan fragments 
 
 0.57 
 
 0.005 
 
 0.62 
 
 0.81 
 
 0.007 
 
 0.008 
 
 0.22 
 
 0.22 
 
 0.52 
 
 0.072 
 
THERMAL EXPANSION OF LIMESTONES AND DOLOMITES 
 
 TABLE 2 - Continued 
 
 Sample 
 no. 
 
 Location 
 (nearest town) 
 
 Formation- 1 
 member 
 
 Description 
 
 Median 
 grain 
 size (mm) 
 
 DOLOMITES 
 
 63-48 Brooklyn, Q Salem 
 
 65-36 Osborne, Q Racine 
 
 65-4 1A Kankakee, Q Racine 
 (lower beds) 
 
 65-41B Kankakee, Q Racine 
 (upper beds) 
 
 61-6 Manteno, Q Racine 
 
 62-17C Thornton, Q Racine 
 
 61-7 Bourbonnais, Q Waukesha 
 
 65-30 Kankakee, Q 
 (lowermost 
 beds) 
 
 62-16A Joliet, Q 
 
 (upper level) 
 
 65-40A Joliet, Q 
 
 (lowest level) 
 
 65-38B Rockton, Q 
 
 (upper 6 ft.) 
 
 65-38C Rockton, Q 
 (middle, 3- 
 15 ft.) 
 
 65-38A Rockton, Q 
 
 (lowest 3 ft.) 
 
 65-39 Rockford, Q 
 
 Joliet 
 
 Joliet 
 
 Kankakee 
 
 Dunleith 
 Fairplay 
 
 Dunleith 
 
 Eagle Point 
 
 Fine-grained mosaic, micro- 
 porous, brown 
 
 Medium— grained mosaic, macro- 
 porous, gray, reef type 
 
 Fine-grained mosaic, macro- 
 porous, gray, inter-reef 
 type 
 
 Medium— grained mosaic, non- 
 porous, buff, inter-reef type 
 
 Medium-grained mosaic, mac- 
 roporous, mottled gray and 
 buff, reef type 
 
 Fine-grained mosaic, macro- 
 porous, gray, reef type 
 
 Fine-grained mosaic, macro- 
 and microporous, buff, non- 
 reef type 
 
 Fine-grained mosaic, macro- 
 porous, light gray with pink 
 mottling, nonreef type 
 
 Fine-grained mosaic, thinly 
 laminated, nonporous, gray 
 
 Fine grained, few clay part- 
 ings; light gray nonreef type 
 
 Coarse-grained mosaic, macro- 
 porous with patches of fine- 
 grained material 
 
 Same as 65-38B; chert present 
 in adjacent beds 
 
 0.06 
 
 0.10 
 
 0.05 
 
 Dunleith Same as 65-38B; chert-face 
 Eagle Point beds 
 
 0.07 
 
 0.02 
 
 0.03 
 
 0.01 
 
 0.03 
 
 0.02 
 
 0.38 
 
 0.39 
 
 0.35 
 
 Dunleith Coarse-grained mosaic, some 0.34 
 Fairplay (?) fine-grained material, macro- 
 porous, mottled buff 
 
10 ILLINOIS STATE GEOLOGICAL SURVEY CIRCULAR 415 
 
 TABLE 2 - Continued 
 
 Sample 
 no. 
 
 Location 
 (nearest town) 
 
 Formation 
 member 
 
 Description 
 
 Median 
 
 grain 
 
 size2(mm) 
 
 DOLOMITES (cont.) 
 
 65-37A Polo, Q 
 
 (10-15 ft. 
 above 
 Q floor) 
 
 Nachusa 
 
 Interlayered fine- and 
 medium— grained mosaics, with 
 globular structure, dense, 
 brownish gray 
 
 0.05 
 
 DOLOMITIC LIMESTONES 
 
 65-37C Polo, Q Nachusa 
 (lowest 5 ft.) 
 
 65-37B Polo, Q 
 
 (5-10 ft. 
 above 
 Q floor) 
 
 65-31 Lowell, Otc. 
 
 Nachusa 
 
 Quimbys Mill 
 
 Fine grained, interlayered 
 with medium-grained mosaic, 
 globular structure, gray 
 
 Very fine grained with 
 medium-grained sparry 
 calcite fossil fragments, 
 dense, mottled brown 
 and gray 
 
 Interbedded fine-grained 
 calcite and medium-grained 
 dolomite mosaic, dense 
 
 0.06 
 
 0.02 
 
 0.01 
 
 1-The samples are listed within rock categories from youngest to oldest geologic 
 
 formation. 
 ^Crystalline particle size. 
 
 Q = quarry or mine. 
 \f5464 indicates well core from 5464 ft. below surface. 
 ^Otc. = outcrop. 
 
 The density of each specimen was determined from direct measurements of 
 the dimensions and weights of the specimen cores. 
 
 EXPERIMENTAL METHODS 
 
 Description of Apparatus 
 
 The apparatus used in this study is similar to that described in the ASTM 
 test for linear thermal expansion of rigid solids with a vitreous silica dilatometer 
 (ASTM, 1963). The dilatometer (fig. 3) consists of a vitreous silica tube, closed 
 at one end, in which a specimen, 1 inch in diameter and 4 inches long, is placed. 
 A vitreous silica push rod rests on top of the specimen and actuates the dial in- 
 dicator attached to the outer tube by means of a bracket made of very low expan- 
 sive metal (invar). A dial indicator was used that has indications at each 0.00005 
 inch and that is capable of indicating length changes as little as 0.00001 inch. 
 The temperature of the specimen was measured by using an iron-constantan ther- 
 mocouple that rested on the top edge of the specimen. The emf -temperature table 
 of Corruccini and Shenker (1953) was used to obtain the temperature. 
 
THERMAL EXPANSION OF LIMESTONES AND DOLOMITES 
 TABLE 3 - MINERAL ANALYSES OF SAMPLES 1 
 
 11 
 
 
 LIMESTONES 
 
 Sample 
 
 DOLOMITES 
 
 Sample 
 
 Calcite 
 
 Dolomite 
 
 Quartz & 
 
 Calcite 
 
 Dolomite 
 
 Quartz & 
 
 no. 
 
 a) 2 
 
 (%) 
 
 clay (%) 
 
 no. 
 
 (%) 
 
 (%) 2 
 
 clay (%) 
 
 65-33A 
 
 99.0 
 
 ND 
 
 1.0 
 
 61-6 
 
 ND C 
 
 99.2 
 
 0.8 
 
 62-22 
 
 98.5 
 
 ND 
 
 1.5 
 
 61-17C 
 
 ND 
 
 99.2 
 
 0.8 
 
 JN-12 
 
 98.4 
 
 ND 
 
 1.6 
 
 65-41B 
 
 ND 
 
 99.1 
 
 0.9 
 
 65-1 
 
 98.3 
 
 ND 
 
 1.7 
 
 65-36 
 
 Tr 4 
 
 99.0 
 
 1.0 
 
 63-39 
 
 98.1 
 
 ND 
 
 1.9 
 
 65-38B 
 
 ND 
 
 98.7 
 
 1.3 
 
 63-40 
 
 97.8 
 
 ND 
 
 2.2 
 
 65-38A 
 
 ND 
 
 98.6 
 
 1.4 
 
 64-12 
 
 97.4 
 
 ND 
 
 2.6 
 
 65-39 
 
 ND 
 
 98.2 
 
 1.8 
 
 62-24M 
 
 97.0 
 
 ND 
 
 3.0 
 
 65-38C 
 
 ND 
 
 98.1 
 
 1.9 
 
 RA-7 
 
 96.7 
 
 2 
 
 1.3 
 
 65-41A 
 
 2 
 
 96.5 
 
 1.5 
 
 K-14 
 
 96.1 
 
 ND 
 
 3.9 
 
 63-48 
 
 5 
 
 92.5 
 
 2.5 
 
 63-41 
 
 96.0 
 
 3 
 
 1.0 
 
 61-16A 
 
 ND 
 
 90.0 
 
 10.0 
 
 65-42 
 
 95.7 
 
 3 
 
 1.3 
 
 61-7 
 
 ND 
 
 90.0 
 
 10.0 
 
 62-26 
 
 95.6 
 
 3 
 
 1.4 
 
 65-30 
 
 ND 
 
 89.0 
 
 11.0 
 
 62-25 
 
 95.0 
 
 4 
 
 0.6 
 
 65-40A 
 
 ND 
 
 88.7 
 
 11.3 
 
 65-32L 
 
 94.1 
 
 3 
 
 2.9 
 
 65-37A 
 
 5 
 
 85.9 
 
 9.1 
 
 C2740 
 
 93.2 
 
 3 
 
 3.8 
 
 
 
 
 
 C4774 
 
 93.1 
 
 6 
 
 0.9 
 
 
 
 
 
 62-27 
 
 86.6 
 
 10 
 
 3.4 
 
 
 DOLOMITIC LIMESTONES 
 
 63-42 
 
 86.2 
 
 12 
 
 1.8 
 
 65-3 7B 
 
 70 
 
 20.9 
 
 9.1 
 
 65-32U 
 
 83.7 
 
 13 
 
 3.3 
 
 65-37C 
 
 55 
 
 35.3 
 
 9.7 
 
 62-24 
 
 77.9 
 
 9 
 
 13.1 
 
 65-31 
 
 60 
 
 35.3 
 
 2.7 
 
 Based on X-ray analysis and insoluble residue. The quartz and clay columns also 
 include any organic material, chert, pyrite, and other HC1 insoluble minerals 
 that may be present in trace amounts. 
 
 By difference. 
 
 Not detected by X-ray analysis. 
 
 Trace amount, 1 percent or less. 
 
12 
 
 ILLINOIS STATE GEOLOGICAL SURVEY CIRCULAR 415 
 
 Diol indicator 
 
 Invar post 
 
 Thermocouple 
 
 Vitreous silica tube 
 and push rod 
 
 The end of the dilatometer con- 
 taining the specimen was immersed in 
 an ethylene-glycol bath whose temper- 
 ature could be controlled in the range 
 -4 to 180° E. A bracket, attached to 
 the outer tube, rested on an insulated 
 lid that was placed over the bath, and 
 thus the dial indicator and its attach- 
 ments remained essentially at room tem- 
 perature throughout the test. The bath 
 was stirred and cooled by running re- 
 frigerated ethylene-glycol water through 
 copper tubing in the bath and was heat- 
 ed by a 250 watt knife heater. 
 
 Test Procedures 
 
 The length and test specimen 
 was measured with a micrometer to the 
 nearest 0. 0001 inch at room tempera- 
 ture. The dilatometer with a specimen 
 was assembled and the specimen end 
 of the apparatus was lowered into the 
 bath. The specimen temperature was 
 lowered to -4° F. for approximately 12 
 hours prior to heating. The specimen 
 temperature was increased and dial in- 
 dicator and emf readings were taken 
 every 1 to lj hours until the tempera- 
 ture reached 176° F. or slightly above. 
 The maximum temperature was main- 
 tained until an equilibrium length of 
 the specimen was obtained (less than 
 5 minutes). The bath and dilatometer 
 were cooled to less than 100° F. be- 
 fore a new specimen was placed in the 
 dilatometer for testing. 
 
 For practical reasons, no at- 
 tempt was made to hold the temperature 
 constant at regular intervals during the 
 heating phase. The effect of such non- 
 equilibrium conditions was minimized 
 by maintaining a very slow rate of heat- 
 ing (0. 10 to 0. 11° F. per minute) and a uniform heat distribution in the bath. The 
 rate of expansion and heating was slow enough so that there was no indication of 
 measurable expansion between the time of taking the emf and dial indicator data. 
 
 Ther mocouple 
 
 Specimen 
 
 Figure 3 - Diagram of a dilatometer, an 
 expansion measuring apparatus. 
 
 Adjustment and Accuracy of Data 
 
 Inspection of figure 3 indicates that the dial indicator gives the difference 
 between the expansion of the specimen and that of a length of the vitreous silica 
 
THERMAL EXPANSION OF LIMESTONES AND DOLOMITES 13 
 
 tube equal to the length of the specimen. The expansivity of vitreous silica from 
 32 to 176° F. is 0.270 x 10" 6 (calculated from data of Beattie et al. , 1941, table 
 III, p. 377), and this value was used to adjust the measurements to give the ex- 
 pansion of the rock cylinders. Recent measurements of Oldfield (1964) indicate 
 that normal commercial grades of good quality vitreous silica conform very close- 
 ly in expansion within the limits of error of ±0.054 x 10""/° F. 
 
 Values of the unit expansion were computed for each measurement of tem- 
 perature and expansion and were plotted on graph paper. The expansivity within 
 each of the temperature ranges of -4 to 68° F. , -4 to 140° F. , and 68 to 140° F. 
 were calculated (equation 4) for each test specimen with the aid of a smoothed 
 curve drawn through the experimental points. 
 
 The accuracy of the measurements of expansion, temperature, and speci- 
 men length are such that for a temperature interval of 35° F. , the maximum and 
 minimum values of the expansivity agree with that determined to within ±0. 1 x 
 10" 6. To check the results, expansion tests were made on materials of known 
 expansion and the results are in good agreement (table 4). Furthermore, there 
 was no measurable lag in temperature between the thermocouple measurement and 
 the predicted expansion of the oxygen-free high conductivity (OFHC) copper. Be- 
 cause the conductivity of copper is greater than that of limestone, there probably 
 was a small lag in the average temperature of the specimen from that indicated 
 by the thermocouple. The use of 0.270 x 10" ^ for the expansivity of the vitreous 
 silica tube is justified by the observed expansion of the copper and other speci- 
 mens of known expansion listed in table 4. 
 
 RESULTS OF EXPANSION TESTS 
 
 Remarkably good agreement was obtained between the two specimens of 
 like orientation from the more homogeneous limestone and dolomite sample blocks. 
 The maximum difference in expansivity in the range -4 to 140° F. between the two 
 
 TABLE 4 - COMPARISON OF EXPANSION AS MEASURED IN 
 OTHER LABORATORIES WITH THAT REPORTED IN THIS INVESTIGATION 
 
 Sample 
 
 Expansivity 
 
 Other labs, 
 
 This study 
 
 Temp. 
 
 interval 
 
 (°F.) 
 
 NBS, JL J 
 
 OFHC copper2 
 Indiana Ls. 
 
 3.16 x 10 
 
 9.2 
 2.8 
 
 -6 
 
 -4 to 140 
 -4 to 140 
 -4 to 140 
 -4 to 140 
 68 to 121 
 
 Sample and expansion data courtesy of U. S. National Bureau of Standards 
 (R. K. Kirby, personal communication). 
 ^Expansion data from ASTM (1963). 
 Souder and Hidnert (1919) . The sample for this study came from the same 
 geologic formation as the one tested by Souder and Hidnert. 
 
14 ILLINOIS STATE GEOLOGICAL SURVEY CIRCULAR 415 
 
 specimens of like orientation from the same sample was 0.5 x 10"°/° F.; the av- 
 erage for all samples was 0.2 x 10"°/ ° F. 
 
 The average curve of expansion of the four specimens of each of several 
 limestones is shown in figure 4. The corresponding sample numbers are indicated 
 adjacent to the curves. The average expansivity of the limestone specimens de- 
 termined from expansion curves over three different temperature ranges are listed 
 in table 5. Average values of the expansivity measured parallel to the bedding 
 and at right angles to bedding of the stone are given in table 5. 
 
 Typical dolomite expansion curves are shown in figure 5, and average ex- 
 pansivity values of each of the dolomites and dolomitic limestones are given in 
 table 6. 
 
 The average expansivity of the limestones within the temperature range of 
 -4 to 68° F. is 1.9 x 10" 6 /° F. ; within the range 68 to 176° F. it is 4.5 x 10" 6 /° J 
 The corresponding average expansivities of the dolomites are 4.3 x 10 -6 /° F. and 
 5.5 x 10"v° F., respectively. Thus, generally, dolomites expand over two times 
 more than limestones in the range -4 to 68° F. The expansivities, in all cases, 
 are greater in the range 68 to 176° F. than in the lower temperature range. This 
 difference is greater in some samples than in others. 
 
 DISCUSSION OF RESULTS 
 
 Effect of Bedding Orientation 
 
 Expansivities measured perpendicular to the bedding differ markedly from 
 those measured parallel in certain samples. The effect of the rock orientation on 
 expansion is shown by the curves in figure 6. These expansion curves illustrate 
 four samples in which the effect was the most pronounced. Expansion curves are 
 shown for each of the specimens taken from a sample of Kimmswick Limestone 
 (63-42). The expansivities calculated from the two perpendicular specimens of 
 the sample between -4 and 140° F. are 6. 6 x 10" 6 and 7. 1 x 10" 6 and from the 
 parallel specimens 4.5 x 10"° and 4.7 x 10". It is interesting to note that one 
 of the parallel oriented specimens actually has a negative expansion between 
 about 32 and 75° F. The expansion curves of the other limestones and the dolo- 
 mite shown in figure 6 are based on the average expansion of the two specimens 
 of like orientation. 
 
 Texturally, the Kimmswick and Harrodsburg are both coarse-grained cri- 
 noidal limestones, and the St. Louis is a very fine-grained limestone. It is of 
 interest that the sample of St. Louis Limestone expanded more parallel than per- 
 pendicular to its bedding, which is opposite to the expansion of the coarse-grained 
 limestones. The direction of greatest expansion for each of the samples tested 
 is given in table 7. In a few samples, the observed differences are in the second 
 decimal and thus are not evident in tables 5 and 6. 
 
 The effect of anisotropic expansion of coarse aggregates on the durability 
 of concrete containing such particles is a matter of little concern at present. How- 
 ever, because certain limestones show a pronounced difference in expansivity rel- 
 ative to the bedding, the effect may be of some consequence. 
 
 Consideration of the anisotropic expansion of calcite crystals referred to 
 earlier accounts for the anistropic expansion of pure limestones. Within the 
 temperature range of 64 to 122° F., calcite crystals have an expansivity near 
 
THERMAL EXPANSION OF LIMESTONES AND DOLOMITES 15 
 
 Kimmswick 
 (63- 41) 
 
 Moccasin Springs 
 (RA-7) 
 
 Burlington 
 (62-26) 
 
 o 
 
 
 o 
 
 c 
 
 i 
 
 o 
 
 Kimmswick 
 (C4774) 
 
 Salem 
 
 (63-39) 
 
 Livingston 
 
 (65-1) 
 
 St. Clair 
 
 (64-12) 
 
 Wapsipinicon 
 (65-33) 
 
 -4 14 32 50 66 86 104 122 140 158 176 
 
 Temperature (°F) 
 
 Figure 4 - Expansion curves of limestones— average of four specimens. (Each curve 
 
 has aAL value of zero at -4° F.) 
 
16 ILLINOIS STATE GEOLOGICAL SURVEY CIRCULAR 415 
 
 EXPLANATION OF PLATE 1 
 
 FIGURE 
 
 1 Kimmswick Limestone (63-41). Dusty crinoid fragments with light 
 colored secondary calcite overgrowths surrounded by dark, fine- 
 grained, distorted bryozoan fragments and scattered microspar grains, 
 Plane polarized light. 
 
 2 Harrodsburg Member of Ullin Limestone (62-25). Crinoid fragments 
 with secondary calcite overgrowths interlayered with dark gray very 
 fine-grained bryozoan fragments. Plane polarized light. 
 
 3 "Quincy bed" of the Burlington Limestone (62-26). Very large cri- 
 noid fragments surrounded by small dark colored fragments of very 
 fine-grained bryozoans and a few grains of microspar. Plane po- 
 larized light. 
 
 EXPLANATION OF PLATE 2 
 
 FIGURE 
 
 1 Rocher Member of Salem Limestone (63-39). Dark colored very fine- 
 grained fossil fragments (ostracodes, endothyrids, bryozoans) ce- 
 mented by sparry calcite with a few large crinoid grains. Plane 
 polarized light. 
 
 2 Livingston Limestone (65-1). Pseudobreccia, very fine-grained and 
 dark algal calcite with patches of sparry calcite of grain size near 
 0.040 mm. The sparry calcite is a product of recrystallization of 
 the fine-grained material. Plane polarized light. 
 
 3 Fredonia Member of the Ste. Genevieve Limestone (62-22). Oolites 
 (medium- to coarse-grained calcite surrounded by very fine-grained 
 calcite) cemented by sparry calcite of grain size near 0.02 mm. 
 Plane polarized light. 
 
 EXPLANATION OF PLATE 3 
 
 FIGURE 
 
 1 Kinkaid Limestone (62-24). Few large calcite grains surrounded by 
 very fine-grained calcite. A few black pyrite grains present. Plane 
 polarized light. 
 
 2 Wapsipinicon Limestone (65-33). Very fine-grained mosaic of cal- 
 cite particles. The black areas are also calcite. Crossed polarized 
 light. The small circles and dots appearing within some grains are 
 due to imperfections in the plastic slide support. 
 
Illinois State Geological Survey 
 
 Circular 415 — Plate 1 
 
Illinois State Geological Survey 
 
 Circular 415 — Plate 2 
 
 
 1/2 mm 
 
 
 , 1/2 mm , 
 
 /2 mm 
 
Illinois State Geological Survey 
 
 Circular 415 — Plate 3 
 
 1/2 mm 
 
 
 
Illinois State Geological Survey 
 
 Circular 415 — Plate 4 
 
 «*fW% 
 
 
 
 V. 
 
 ft 
 
THERMAL EXPANSION OF LIMESTONES AND DOLOMITES 21 
 
 15 x 10"° parallel to the c-axis of the crystal and approximately -3 x 10" D per- 
 pendicular to the c-axis. Thus, if the calcite particles were preferentially ori- 
 ented in the limestone, the direction in which the c-axes were oriented would be 
 the direction of a greatest expansion. Rosenholtz and Smith (1951) measured the 
 thermal expansion of specimens of Yule marble and compared their results with 
 the known direction of preferred orientation of the calcite particles in this marble 
 as determined by Turner (1949). Their comparison verified that preferred orienta- 
 tion of anisotropic crystalline grains in rocks affects the thermal expansion in a 
 systematic manner. 
 
 The expansion curves in figure 6 strongly suggest that there is some de- 
 gree of systematic preferred orientation of the calcite particles in these limestones. 
 Each of the coarse-grained crinoidal limestones tested, particularly the Kimms- 
 wick Limestone (samples 63-42 and 65-42) and the Harrodsburg Member of the 
 Ullin Formation (sample 62-25), show greater expansion perpendicular to the bed- 
 ding than parallel. It is therefore postulated that the c-axes of the grains in 
 these limestones are mainly oriented near the vertical direction. The samples 
 that expanded more parallel to the bedding than perpendicular (table 7) are all 
 limestones that consist predominantly of fine grains (median grain size less than 
 0.050 mm). Fine-grained limestones that are not in this category are the Kinkaid 
 and Salem (Rocher Member) Limestones. The sample of Kinkaid contains numer- 
 ous flakes of clay minerals, which are themselves preferentially oriented in the 
 rock, causing a greater expansion perpendicular to the bedding. However, no 
 explanation accounts for the anomalous expansion of the sample of the Salem 
 (63-39). Nevertheless, seven out of nine predominantly fine-grained limestones 
 expand somewhat more parallel to the bedding, which suggests a slight preferred 
 orientation of the very fine-grained calcite particles (micrite) in these limestones 
 with c-axes tending to be parallel to the bedding. 
 
 Each of the dolomite samples had an average expansion perpendicular to 
 the bedding greater than that parallel to the bedding. Only six of the individual 
 test specimens taken parallel to bedding expanded more than either of the two 
 perpendicular specimens-a total of fifty different comparisons with only six ex- 
 ceptions. C- and a-axes expansivities of dolomite single crystals are not known, 
 but it is probable, considering the similarities of crystal structure of calcite and 
 dolomite, that the expansivity of dolomite crystals in the c-axis direction is great- 
 er than at right angles to the c-axes. Given this assumption, it follows that the 
 
 EXPLANATION OF PLATE 4 
 
 FIGURE 
 
 1 Kankakee Dolomite (65-40A). Non-reef type; equant mosaic of in- 
 dividuals; median grain size of 0.02 mm. Crossed polarized light. 
 
 2 Racine Dolomite (65-3 6). Reef type; equant mosaic of individuals; 
 median grain size of 0.10 mm. Crossed polarized light. 
 
 3 Eagle Point Member of Dunleith Dolomite (65-38A). Non-reef type; 
 equant mosaic of individuals; median grain size 0.35 mm. Crossed 
 polarized light. 
 
22 ILLINOIS STATE GEOLOGICAL SURVEY CIRCULAR 415 
 
 TABLE 5 - THERMAL EXPANSION OF ILLINOIS LIMESTONE 
 
 
 
 
 
 Average 
 
 expansivity 
 
 
 Median 
 
 Sample 
 
 Specimen 
 
 
 Temperature 
 
 interval 
 
 (°F.) 
 
 
 grain 
 
 no. 
 
 attitude 
 
 -4 to 
 
 140° 
 
 -4 to 68° 
 
 68 to 
 
 176° 
 
 sizel(mm) 
 
 JN-12 
 
 1 
 II 
 
 2.2 x 
 2.2 
 
 10~ 6 
 
 1.9 
 2.1 
 
 x 10" 6 
 
 2.6 x 
 
 2.6 
 
 io- 6 
 
 0.010 
 
 65-1 
 
 1 
 II 
 
 2.3 
 
 
 2.1 
 
 
 2.8 
 
 
 0.010 
 
 62-24M 
 
 1 
 II 
 
 2.4 
 
 
 2.2 
 
 
 3.2 
 
 
 0.020 
 
 62-24 
 
 1 
 II 
 
 3.5 
 3.2 
 
 
 3.2 
 2.9 
 
 
 4.1 
 3.7 
 
 
 0.010 
 
 62-22 
 
 1 
 II 
 
 1.8 
 1.9 
 
 
 1.6 
 1.8 
 
 
 2.7 
 2.5 
 
 
 0.010 
 
 K-14 
 
 1 
 II 
 
 2.0 
 2.5 
 
 
 1.8 
 2.3 
 
 
 2.4 
 2.9 
 
 
 0.003 
 
 63-39 
 
 1 
 
 II 
 
 2.2 
 2.1 
 
 
 2.0 
 1.8 
 
 
 2.6 
 2.8 
 
 
 0.030 
 
 63-40 
 
 1 
 II 
 
 2.2 
 2.4 
 
 
 1.9 
 2.2 
 
 
 2.6 
 
 2.5 
 
 
 0.040 
 
 62-25 
 
 1 
 II 
 
 2.2 
 1.5 
 
 
 1.3 
 
 0.8 
 
 
 5.1 
 3.1 
 
 
 0.060 
 
 62-26 
 
 1 
 II 
 
 2.3 
 2.2 
 
 
 1.0 
 0.8 
 
 
 5.8 
 5.9 
 
 
 0.22 
 
 65-32U 
 
 65-32L 
 
 65-33 
 
 C2740 
 
 ± 
 II 
 
 1 
 II 
 
 1 
 
 II 
 
 1 
 
 4.7 
 
 3.2 
 3.1 
 
 2.1 
 2.2 
 
 3.6 
 
 2.4 
 
 1.9 
 1.7 
 
 2.0 
 2.1 
 
 2.6 
 
 9.8 
 
 6.9 
 6.9 
 
 2.2 
 2.3 
 
 5.9 
 
 0.54 
 
 0.57 
 
 0.005 
 
 0.62 
 
 RA-7 
 
 ± 
 
 2.6 
 
 1.1 
 
 6.3 
 
 0.81 
 
 64-12 
 
 62-27 
 
 1 
 
 1 
 II 
 
 2.3 
 
 2.3 
 
 2.4 
 2.4 
 
 2.0 
 2.1 
 
 2.3 
 2.3 
 
 2.9 
 2.8 
 
 2.7 
 2.7 
 
 0.007 
 
 0.008 
 
THERMAL EXPANSION OF LIMESTONES AND DOLOMITES 23 
 
 TABLE 5 - Continued 
 
 
 Specimen 
 attitude 
 
 
 
 
 Average 
 
 expansivity 
 
 
 
 Median 
 
 Sample 
 no. 
 
 -4 to 
 
 Temperature interval 
 140° -4 to 68° 
 
 (°F.) 
 68 to 
 
 176° 
 
 grain 
 size-'-(mm) 
 
 63-41 
 
 1 
 II 
 
 3.3 
 3.2 
 
 X 
 
 icr 6 
 
 1.5 
 
 1.6 
 
 X 
 
 10~ 6 
 
 7.4 
 6.4 
 
 X 
 
 10" 6 
 
 0.22 
 
 63-42 
 
 1 
 II 
 
 3.8 
 2.6 
 
 
 
 1.7 
 
 0.6 
 
 
 
 8.2 
 7.1 
 
 
 
 0.22 
 
 65-42 
 
 1 
 II 
 
 3.5 
 2.7 
 
 
 
 2.3 
 1.5 
 
 
 
 6.1 
 5.3 
 
 
 
 0.52 
 
 C4774 
 
 1 
 
 II 
 
 1.5 
 
 
 
 0.8 
 
 
 
 3.6 
 
 
 
 0.072 
 
 Crystalline particle size. 
 
 preferred orientation of the crystalline particles in the dolomites tested is such 
 that the c-axes tend to be oriented at right angles to the bedding, or, in other 
 words, vertical. Although the degree intensity of preferred orientation is prob- 
 ably small, it does suggest the possibility that the overburden pressure that ex- 
 isted on the rocks during dolomitization influenced the orientation of the second- 
 ary dolomite grains. On the other hand, because many of the dolomites tested 
 were originally coarser varieties of limestones, which probably contain preferen- 
 tially oriented particles, as postulated for the coarse-grained limestones tested 
 in this study, the orientation of the secondary dolomite grains may merely reflect 
 the original limestone texture (A. V. Carozzi, personal communication). 
 
 Grain-Size Effects 
 
 The nonlinear expansion of certain limestones (the upper four curves in 
 fig. 4 and the expansion curve of the Dunleith Dolomite in fig. 5) is striking. A 
 measure of the degree of nonlinearity of the expansion curve is given by the dif- 
 ference between the expansivities calculated for the -4 to 68° F. range and that 
 for the 68 to 176° F. range (expansivity difference). The average expansivity dif- 
 ference for each sample is plotted against the median grain size in figure 7. Con- 
 siderable scatter is evident in the coarse grain sizes, but a definite correlation 
 is seen between the expansivity difference and grain size. The degree of nonlin- 
 earity, or expansivity difference, of the limestones increases with increasing 
 grain size, as shown by the solid dots in the figure. The effect is not as pro- 
 nounced in the dolomites tested, although the same trend is evident from the dis- 
 tribution of the data. 
 
 A comparison of the microscopic texture of the limestones and the type of 
 expansion curve suggests that the type of grain boundary influences the character 
 of the expansion curve. Nonlinear expansion occurs in limestones containing nu- 
 merous large grains in direct contact with one another (pi. 1), whereas essen- 
 tially linear expansion occurs in limestones with large calcite grains, scattered 
 throughout the rock, which are surrounded by very fine-grained particles (pi. 2; 
 
24 ILLINOIS STATE GEOLOGICAL SURVEY CIRCULAR 415 
 
 Racine 
 (65-36) 
 
 Joliet 
 (62-I6A) 
 
 Nachusa 
 (65-37A) 
 
 Kankakee 
 (65-40A) 
 
 Waukesha 
 (61-7) 
 
 Salem 
 (63-48) 
 Dunleith 
 (65-38A) 
 
 Quimbys Mill 
 (65-31) 
 
 ■4 14 32 50 66 86 104 122 140 158 176 
 
 Temperature (°F) 
 
 Figure 5 - Expansion curves of dolomites and a dolomitic limestone (Quimbys Mill, 
 65-31). (Each curve has a AL value of zero at -4° F. ) 
 
 L 
 
THERMAL EXPANSION OF LIMESTONES AND DOLOMITES 25 
 
 TABLE 6 - THERMAL EXPANSION OF DOLOMITES AND DOLOMITIC LIMESTONES 
 
 Sample 
 no. 
 
 Specimen 
 attitude 
 
 Average expansivity 
 
 Temperature interval (°F.) 
 -4 to 140° -4 to 68° 68 to 176 { 
 
 Median 
 
 grain 
 
 sizel(mm) 
 
 63-48 
 
 65-36 
 
 65-41A 
 
 65-41B 
 
 61-6 
 
 62-17C 
 
 61-7 
 
 65-30 
 
 62-16A 
 
 65-40A 
 
 65-38B 
 
 65-38C 
 
 65-38A 
 
 65-39 
 
 65-37A 
 
 DOLOMITES 
 
 L 
 II 
 
 4.4 x 10"° 
 
 4.2 x 10-° 
 
 5.0 
 
 1 
 
 5.0 
 
 4.8 
 
 5.4 
 
 II 
 
 4.8 
 
 4.7 
 
 5.3 
 
 1 
 
 4.8 
 
 4.7 
 
 5.3 
 
 II 
 
 4.6 
 
 4.4 
 
 5.0 
 
 1 
 
 4.7 
 
 4.3 
 
 5.5 
 
 II 
 
 4.6 
 
 4.3 
 
 5.3 
 
 1 
 II 
 
 4.8 
 
 4.4 
 
 5.5 
 
 1 
 II 
 
 4.7 
 
 4.5 
 
 5.2 
 
 ± 
 
 4.6 
 
 4.2 
 
 5.7 
 
 II 
 
 4.4 
 
 4.0 
 
 5.5 
 
 1 
 
 4.7 
 
 4.3 
 
 5.5 
 
 II 
 
 4.6 
 
 4.2 
 
 5.4 
 
 1 
 
 5.0 
 
 4.7 
 
 5.9 
 
 II 
 
 4.4 
 
 4.4 
 
 5.7 
 
 1 
 
 5.0 
 
 4.7 
 
 5.5 
 
 II 
 
 4.8 
 
 4.6 
 
 5.2 
 
 1 
 
 4.2 
 
 3.6 
 
 5.6 
 
 II 
 
 4.1 
 
 3.5 
 
 5.3 
 
 1 
 
 4.4 
 
 3.8 
 
 5.8 
 
 II 
 
 4.2 
 
 3.6 
 
 5.5 
 
 1 
 
 4.34 
 
 3.6 
 
 6.1 
 
 II 
 
 4.31 
 
 3.6 
 
 5.8 
 
 1 
 
 4.5 
 
 4.1 
 
 5.2 
 
 II 
 
 4.4 
 
 4.0 
 
 5.5 
 
 1 
 
 5.0 
 
 4.7 
 
 5.7 
 
 II 
 
 4.5 
 
 4.3 
 
 5.0 
 
 .06 
 .10 
 .05 
 .07 
 .07 
 .02 
 .03 
 .01 
 .03 
 .02 
 .38 
 .39 
 .35 
 .34 
 .06 
 
26 ILLINOIS STATE GEOLOGICAL SURVEY CIRCULAR 415 
 
 TABLE 6 - Continued 
 
 Sample 
 no. 
 
 Specimen 
 attitude 
 
 Average expansivity 
 
 Temperature interval (°F.) 
 -4 to 140° -4 to 68° 68 to 176' 
 
 Median 
 
 grain 
 
 size-Hmm) 
 
 DOLOMITIC LIMESTONES 
 
 65-37C 
 
 65-37B 
 
 65-31 
 
 1 
 
 II 
 
 2.9 x 
 
 io- 
 
 -6 
 
 2.6 x 10- 6 
 
 3.7 
 
 1 
 
 2.7 
 
 
 
 2.5 
 
 3.1 
 
 II 
 
 2.6 
 
 
 
 2.3 
 
 3.1 
 
 1 
 
 3.5 
 
 
 
 3.2 
 
 4.0 
 
 II 
 
 3.1 
 
 
 
 2.9 
 
 3.5 
 
 .06 
 
 ,02 
 
 .01 
 
 "Crystalline particle size. 
 
 pi. 3, fig. 2). This possibly indicates that thermal stresses are more directly 
 transmitted across grain boundaries between very fine-grained calcite than be- 
 tween large calcite particles. The dolomites (pi. 4) are equant grain mosaics 
 in areas as small as that viewed with a microscope, yet to the naked eye, they 
 also show varying degrees of inhomogeneity, mainly in grain size and porosity. 
 The porosity, whether macro- or microporosity, does not correlate with the ob- 
 served expansion characteristics of either limestones or dolomites, as does the 
 size of the grains shown in figure 7. 
 
 It is not known why the expansivity of coarse-grained limestones changes 
 so rapidly in the range 60 to 80° F. The reason possibly lies in the fact that dis- 
 torting strain waves created in crystals by thermal energy are affected by proper- 
 ties of grain boundaries or certain imperfections in the crystals, such as fluid in- 
 clusions and dislocations, which are themselves related to grain size. 
 
 The pronounced changes in the rate of expansion with temperature of the 
 coarse-grained limestones in the temperature range studied is significant and may 
 have a bearing on the durability of certain concrete structures containing such 
 aggregates. Each of the expansion curves obtained indicates that expansivity 
 increases with increasing temperatures, but the nonlinearity of the fine-grained 
 limestones (median grain size 0.040 mm and less) and most of the dolomites is 
 minor compared to the rapid increase in expansion of the coarse-grained lime- 
 stones (0.072 mm and over). 
 
 Mineralogical Effects 
 
 The effect of mineral composition on the expansion of the rock is shown 
 by a plot of the percentage of dolomite against the average expansivity of the 
 four specimens of each sample determined between -4 and 140° F. (fig. 8). The 
 plot shows that the greater the dolomite content of the rock, the greater the 
 expansion. In figure 8, a comparison is made of the samples based on grain size, 
 
THERMAL EXPANSION OF LIMESTONES AND DOLOMITES 27 
 
 O 
 
 c 
 
 CD 
 
 O 
 
 c 
 
 ro 
 
 i 
 
 o 
 
 14 
 
 32 
 
 50 66 86 104 122 
 
 Temperature (°F) 
 
 Kimmswick 
 (63- 42) 
 
 Nachusa 
 (65-37A) 
 
 Harrodsburg 
 (62-25) 
 
 St. Louis 
 (K-14) 
 
 40 158 176 
 
 Figure 6 - Expansion curves as measured perpendicular (1) and parallel (||) to the 
 bedding. (Each curve has a AL value of zero at -4° F.) 
 
 L 
 
28 ILLINOIS STATE GEOLOGICAL SURVEY CIRCULAR 415 
 
 The relation between the percentage of dolomite and the expansion is more nar- 
 rowly defined by the fine-grained rocks than the coarse-grained ones. A similar 
 relationship is observed on a plot of the percentage of acid residue (mainly con- 
 sisting of quartz and clay) against expansivity (fig. 9). The points of the fine- 
 grained rocks line up well on a steep sloping line, whereas the coarse-grained 
 ones, which are all rather pure, show a wide variation of expansivity. 
 
 CONCLUSIONS 
 
 The expansion curves of the ten coarse-grained limestones tested show 
 a very low rate of expansion below 60° F. and a much greater rate above 100° F. ; 
 the rate of expansion of the fine-grained samples remains nearly constant. These 
 data show a definite relation between rock texture and the thermal expansion char- 
 acteristics of limestones (and to some extent of dolomites) that heretofore has not 
 been noted. This relation may bear significantly on the use performance as build- 
 ing stones and concrete aggregates. 
 
 Specifically, the data indicate that distinct differences exist in the be- 
 havior of pure limestones and dolomites when they are used in concrete that is 
 subject to climatic conditions ranging from subfreezing to over 100° F. Air tem- 
 peratures are seldom much above 100° F., although when concrete is exposed to 
 direct sunlight in the summer, pavement temperatures higher than this frequent - 
 
 TABLE 7 - THERMAL EXPANSION AND ROCK ORIENTATION 
 
 Greater expansion 
 
 Perpendicular to bedding 
 
 LIMESTONES 
 
 Kinkaid (Buncombe) 
 Salem (Rocher) 
 Ullin (Harrodsburg) 
 Burlington ("Quincy bed") 
 Burlington (Monmouth) 
 Kimmswick (buff, Valmeyer) 
 Kimmswick (gray, Valmeyer) 
 Kimmswick (Thebes) 
 
 All dolomites tested 
 
 DOLOMITES 
 
 Parallel to bedding 
 
 Omega (Omega) 
 Ste. Genevieve (Fredonia) 
 St. Louis (Alton) 
 Salem (Kidd) 
 
 Wapsipinicon (Davenport) 
 St. Clair (Gale) 
 Girardeau (Thebes) 
 
 None 
 
 L Contains 10 percent more insoluble quartz and clay than the other samples 
 The clay present is mainly illite (about 5 to 8 percent) . 
 
THERMAL EXPANSION OF LIMESTONES AND DOLOMITES 29 
 
 
 
 1 
 
 1 1 1 
 
 i 
 
 
 i i 
 
 1 
 
 1 
 
 1 
 
 1 
 
 • 
 
 
 7 
 
 - 
 
 
 
 
 
 
 
 • 
 
 
 
 _ 
 
 6 
 
 - 
 
 • 
 
 Limestone 
 
 
 
 
 
 
 
 
 - 
 
 i 5 
 O 
 
 - 
 
 O 
 
 Dolomite 
 
 
 
 
 
 • 
 • 
 
 
 • 
 
 • 
 
 X 
 
 - 
 
 
 
 
 
 
 
 
 
 
 - 
 
 <->4 
 
 c 
 
 
 
 
 
 
 • 
 • 
 
 
 
 
 • 
 • 
 
 - 
 
 > 
 to 
 c 
 o 
 
 X c 
 
 UJ 
 
 
 
 
 
 
 
 
 
 
 
 8 
 
 
 - 
 
 1 
 
 
 I 
 
 • I 8 
 
 • 
 • 
 i i i 
 
 • 
 
 o 
 o 
 
 • 
 
 08 
 
 o 
 
 • ° 
 
 1 1 
 
 o 
 
 1 
 
 1 
 
 
 
 1 
 
 1 
 
 - 
 
 .002 .004 006 01 02 04 .06 I .2 .4 .6 10 
 
 Median grain size (mm) 
 
 Figure 7 - Effect of grain size on the change of expansivity within the temperature 
 
 range -4 to 176° F. 
 
 ly occur. Such temperature conditions would place coarse-grained limestones 
 in the lower part of the high-expansivity portion of their expansion curve. Dur- 
 ing freezing conditions, these same aggregates, of course, would be on the low- 
 expansivity portion of the curve. Whether or not the nonlinear expansion char- 
 acteristics significantly affect the durability of concrete in such climatic envi- 
 ronments is not known. It is interesting to note, however, that samples show- 
 ing distinctly nonlinear expansion came from five different sources; aggregates 
 from four of these sources have been found unacceptable for use in concrete 
 pavements by the Illinois Division of Highways (1963). Conversely, samples 
 from six different quarries were found to have linear expansion, and aggregates 
 from five of these produce acceptable concrete aggregates. The limestone from 
 the sixth quarry passes the physical tests, but the quarry product includes stra- 
 ta of shale, which is the basis of its rejection. 
 
 Certain limestones, particularly the Harrodsburg and the Kimmswick, and 
 to a lesser extent the Burlington, expand significantly greater perpendicular to 
 
30 
 
 ILLINOIS STATE GEOLOGICAL SURVEY CIRCULAR 415 
 
 
 
 
 
 •{•;•• 
 
 
 
 
 
 
 *°8o 
 
 o 
 
 
 80 
 
 
 
 Med 
 
 on gro 
 
 in size less than 005mm 
 
 - 
 
 
 • 
 
 Med 
 
 an groin size more thon 05mm 
 
 
 1 60 
 
 - 
 
 
 
 
 - 
 
 E 
 o 
 o 
 
 O 40 
 
 - 
 
 
 
 • O 
 
 • 
 
 - 
 
 20 
 
 
 
 
 O 
 
 - 
 
 
 
 • 
 
 • 
 
 O 
 
 • 
 
 ooo 
 
 • 
 
 o 
 
 
 Figure 8 - Effect of dolomite min- 
 eral content on expansion of 
 carbonate rocks. 
 
 2 3 4 5 
 
 Average expansivity (-4 to 140° F) 
 
 6 x I0" 6 
 
 Figure 9 - Effect of acid residue 
 mineral impurities (mainly 
 quartz and clay) on the ex- 
 pansion of limestones. 
 
 Median gram size less than 0.05mm 
 Median grain size more than 05 mm 
 
 20 30 40 
 
 Averoge expansivity (-4tol40°F) 
 
 5 0«I0- 6 
 
 the bedding than parallel. These data suggest that there is a preferred orienta- 
 tion of the calcite particles in these samples and that the c-axes of the grains 
 tend to be oriented at high angles to the bedding planes. 
 
THERMAL EXPANSION OF LIMESTONES AND DOLOMITES 31 
 
 REFERENCES 
 
 Adie, A. J., 1836, On the expansion of different kinds of stone from an increase 
 of temperature with a description of the pyrometer used in making the ex- 
 periments: Royal Soc. Edinburgh Trans. , v. 13, p. 354-372. 
 
 American Society for Testing and Materials, 1963, Tentative method of test for 
 
 linear thermal expansion of rigid solids with a vitreous silica dilatometer: 
 ASTM Designation: E 228-63T, in_1964 Book of ASTM Standards, pt. 30, 
 p. 1-12. 
 
 Austin, J. B., Saini, H., Weigle, J., and Pierce, R. H. H., Jr., 1940, A direct 
 comparison on a crystal of calcite of the X-ray and optical interferometer 
 methods of determining linearly thermal expansion: Phys. Review, ser. 2, 
 v. 57, p. 931-933. 
 
 Bartlett, W. H. C, 1832, Experiments on the expansion and contraction of build- 
 ing stone: Am. Jour. Sci. and Arts, 1st ser., v. 22, p. 136-140. 
 
 Beattie, J. A., Blaisdell, B. E., Kaye, Joseph, Gerry, H. T., and Johnson, C. A., 
 1941, An experimental study of the absolute temperature scale; VIII-The 
 thermal expansion and compressibility of vitreous silica and thermal di- 
 lation of mercury: Am. Acad. Arts and Sci. Proc, v. 74, no. 11, p. 371-36 
 
 Callan, E. J., 1950, The relation of thermal expansion of aggregates to the dura- 
 bility of concrete: U. S. Army Corps of Eng., Vicksburg, Mississippi 
 Waterways Exp. Sta. Bull. 34, 19 p. 
 
 Cook, H. K. , 1956, Thermal properties, in Significance of tests and properties 
 of concrete and concrete aggregates: ASTM Spec. Tech. Pub. 169, 
 p. 325-333. 
 
 Corruccini, R. J., and Shenker, Henry, 1953, Modified 1913 reference tables for 
 iron-constantan thermocouples: U. S. Natl. Bur. Stand. Jour. Research, 
 v. 50, no. 5, p. 229-248. 
 
 Davis, R. E., 1930, A summary of investigations of volume changes in cements, 
 
 mortars, and concretes produced by causes other than stress: ASTM Proc, 
 v. 30, pt. 1, p. 668-685. 
 
 Dunn, J. R. , 1963, Characteristics of various aggregate producing bedrock for- 
 mations in New York State: New York State Department of Public Works, 
 Phys. Research Proj. 4, 258 p. 
 
 Griffith, J. H. , 1936, Thermal expansion of typical American rocks: Iowa Eng. 
 Exp. Sta. Bull. 128, v. 35, no. 19, 36 p. 
 
 Hallock, William, 1891, Preliminary note on the coefficients of thermal expansion 
 of certain rocks: U. S. Geol. Survey Bull. 78, p. 109-118. 
 
 Harvey, R. D. , 1963, Impact resistance of Illinois limestones and dolomites: Il- 
 linois Geol. Survey Circ. 345, 20 p. 
 
 Harvey, R. D. , 1966, Thermal expansion of certain Illinois limestones: Illinois 
 Geol. Survey, Ind. Min. Notes 24, 6 p. 
 
 Harvey, R. D. , in press, Nonlinear thermal expansion of coarse-grained lime- 
 stone: Materials Research and Standards, ASTM. 
 
32 ILLINOIS STATE GEOLOGICAL SURVEY CIRCULAR 415 
 
 Hidnert, Peter, and Souder, Wilmer, 1950, Thermal expansion of solids: U. S. 
 Natl. Bur. Stand. Circ. 486, 29 p. 
 
 Hornebrook, F. B., 1942, A concrete failure attributed to aggregates of low ther- 
 mal coefficient: Am. Concrete Inst. Proc, v. 38, p. 36-43. 
 
 Illinois Division of Highways, 1963, Commercial plants producing aggregates for 
 highway construction in Illinois: Illinois Dept. Public Works and Build- 
 ings, Div. Highways Bull. 23, table 6, p. 25-31. 
 
 Johnson, W. H., and Parsons, W. H. , 1944, Thermal expansion of concrete ag- 
 gregate materials: U. S. Natl. Bur. Stand. Jour. Research, v. 32, no. 3, 
 p. 101-126. 
 
 Koenitzer, L. H. , 1936, Elastic and thermal expansion properties of concrete as 
 
 affected by similar properties of the aggregate: ASTM Proc, v. 36, no. 2, 
 p. 393-410. 
 
 Mather, Bryan, Callan, E. J., Mather, Katharine, and Dodge, N. B., 1953, Lab- 
 oratory investigation of certain limestone aggregates for concrete: U. S. 
 Army Corps ofEng., Vicksburg, Mississippi Waterways Exp. Sta. Tech. 
 Memo. 6-371, 50 p. 
 
 Meyers, S. L. , 1940, Thermal coefficient of expansion of Portland cement: Indus, 
 and Eng. Chemistry, v. 32, p. 1107-1112. 
 
 Meyers, S. L. , 1951, Thermal expansion characteristics of hardened cement paste 
 and concrete: Highway Research Board Proc. , v. 30, p. 193-203. 
 
 Mitchell, L. J., 1956, Thermal properties, in Significance of test and properties 
 of concrete and concrete aggregates: ASTM Spec. Tech. Pub. 169, p. 
 129-135. 
 
 Nisi, H., 1913, Thermal expansion of marble: Math, and Phys. Society (Tokyo) 
 Proc, 2ndser., v. 7, p. 97-102. 
 
 Oldfield, L. F., 1964, Absolute and relative linear thermal expansion coefficients 
 of vitreous silica and platinum: Glass Technology, v. 5, no. 1, p. 41-50. 
 
 Pearson, J. C, 1942, A concrete failure attributed to aggregates of low thermal 
 expansion coefficient: Am. Concrete Inst. Proc, v. 38, p. 29. 
 
 Rosenholtz, J. L. , and Smith, D. T., 1949, Linear thermal expansion of calcite, 
 
 variety iceland spar and Yule marble: Am. Mineralogist, v. 34, p. 846-854. 
 
 Rosenholtz, J. L. , and Smith, D. T., 1951, The directional concentration of optic 
 axes in Yule marble— A comparison of the results of petro-fabric analysis 
 and linear thermal expansion: Am. Jour. Sci., v. 249, no. 5, p. 377-384. 
 
 Souder, W. H., and Hidnert, Peter, 1919, Thermal expansion of insulating ma- 
 terials: U. S. Natl. Bur. Stand., Scient. paper 352, v. 15, p. 387-417. 
 
 Turner, F. J., 1949, Preferred orientation of calcite in Yule marble: Am. Jour. Sci., 
 v. 247, no. 9, p. 593-622. 
 
 Van der Plas, L. , 1962, Preliminary note on the granulometric analysis of sedi- 
 mentary rocks: Sedimentology, v. 1, no. 2, p. 145-157. 
 
 Walker, Stanton, Bloem, D. L. , and Mullen, W. G., 1952, Effects of tempera- 
 ture changes on concrete as influenced by aggregates: Am. Concrete Inst. 
 Proc, v. 48, p. 661-679. 
 
THERMAL EXPANSION OF LIMESTONES AND DOLOMITES 33 
 
 Wheeler, N. E., 1910, On the thermal expansion of rock at high temperatures: 
 Royal Soc. Canada Trans. , ser. 3, v. 4, sec. 3, p. 19-44. 
 
 Willis, T. F., and DeReus, M. E., 1939, Thermal volume change and elasticity 
 
 of aggregates and their effect on concrete: ASTM Proc, v. 39, p. 919-929. 
 
Illinois State Geological Survey Circular 415 
 33 p., 4 pis., 9 figs., 7 tables, 1967 
 
 Printed by Authority of State of Illinois, Ch . 127, IRS, Par. 58.25, 
 
CIRCULAR 415 
 
 ILLINOIS STATE GEOLOGICAL SURVEY 
 
 URBANA agS