US 2141571 A
Abstract available in
Claims available in
Description (OCR text may contain errors)
Examiner '5: cmsmws I m cogwme 0a PL l ir" 2,141,571v 41% Dec. 2 7, 1938.
H. L. KENNEDY ET AL GRINDING OF CEMENT CLINKER Filed Nov. 9, 1935 0 0 0 E M WM n 8 AE WM 5 0 N cc ML 44 NA 5 00 MM .W MM w 3 v MM M Z W Z 6 EOIUUSS NLL L00 M xnuummlw NMDPNEMM: $32 BFPsL cw 2 3456 //N l 7 any 98 6 2 I I I I Reva/afions of Ball Mill Inventors 0 Henry L. Kennedy COMPOSI l lUNS,
Patented Dec. 27, 1938 UNITED STATES Examiner PATENT [OFFICE GRDIDING OF CEMENT CLINKER Henry L. Kennedy, Arlington, and Jacob G. Mark, Cambridge, Mass, assignors to Dewey and Almy Chemical Company, North Cambridge, Mass, a corporation of Massachusetts Application November 9, 1935, Serial No. 49,064
This invention relates to the grinding of cement clinker.
When dry materials are ground in a ball mill, the reduction in particle size usually progresses as a function of mill revolutions until a point is reached known as the limit of free grind where the curve is discontinuous." Thereaftert'al'though the mill may be run for hundreds or perhaps thousands of revolutions, the particle size is not greatly reduced.
The objects of this invention are to increase still further the limit of free grind; to keep the grinding media clean and uncaked form TcTngertiiire thah" measurements-pawns; to increase the output of grinding mills; and to prevent aggregations from occurring in the mill or in the passages leading to the air separators to the end that only unground particles are returned. Other objects are to increase the eflicienc tain recommended cement addition agents by distributing them in a completely un anner throughout the whole mass of cement and to provide means by which extremely small quantities of such agents may be uniformly distributed throughout the whole mill. Other objects will become apparent from the specification.
The limit of free grind effect has been explained by several theories, the most common, and the one usually adhered to by the operators of the mills themselves, is that the balls or grinding media are coated with the pulverized substance; that the blow is cushioned by the coating and, consequently, no further grinding can take place. Indeed, this explanation appears to be true, for if the mill is opened the grinding media almost invariably are found to be coated with a very considerable layer of particles. But this theory by no means entirely explains the effect for, when certain substances are added to the mill, the particle size drops below the limit of free grin although the grinding media may be just as thoroughly coated as before.
If the grinding of cement be taken by way of example, it will be found that under standard grinding practice, a very. considerable proportion of the output of the mill is run back into the mill as oversize rejects from the air separators. Microscopical examination of these rejects shows that there are two components therein; first, small single particles of cement clinker as yet not sufficiently ground, and; second, dense aggregated lumps of clinker, the individual particles of which have been ground to acceptable size.
For a number of years )eeLtallow, rgsin and oleic acid have been used to increase the production of mills. When used, we have noted that the rejects from the air separators contain fewer aggregations of particles, which means, to our minds, that tallow or oleic acid in some manner prevents a certain proportion of aggregates from forming. We have found, in addition, that when substances such as are disclosed by Tucker in U. S. application Serial No. 643,740 filed November 21, 1932 or the Hydroxy-alkyl amines as disclosed by Tucker et al. iiY'U'I'ST Serial No. 756,- 082 filed December 5, 1934 orfiflll n compounds as disclosed by Mark in U. S. application Serial No. 32,906, filed July 24, 1935, are added to the clinker during the process of grinding, much less aggregated material exists in the rejects from the air separator and that the limit of free grind is very materially raised.
The lignin compounds are: li n sul honates, sodium lignate, the sodium salt 0% the condensation product of ligmn and glycerol monochlor- Hydrm, the sodium salt of the condensation product of ligmn an e y ene g yco chlorolignin,
plifiiofderivatfires of l gnin. I
e prepara ion 0 ca cium lignin sulphonates is described in the Howar pa en s, eis'sue 0. 18,268, No. 1,856,558, and No. 1,958,624.
The preparation of glycol lignins is described by Brauns and Hibbert in the Journal of the American Chemical Society, volume 55, page 4720 (1933).
The preparation of phenol lignins is described by Fuchs in the J ourna o e American Chemical Society, volume 58, page 673 (1936).
Additional information, particularly concerning the preparation of co cts of 1i in is found in chapter 3 of The Chemistry of Cellulose and Wood, by Schorger, published by the International Chemical Series in 1926.
By following the methods set forth in this chemical literature, the products listed above may be produced.
These various substances are added to Portland cement to achieve certain effects. A few examples follow:--the substances disclosed in the Tucker application (sole) increase plasticity and workability of the mix. Those disclosed in the application of Tucker et al. materially increase the strength of the hardened concrete.
Speaking generally, these substances have been added to the cement orto the gau ing water by the contractor while the cement is Being mixed on the job, but in some cases it has been the practice to mix the dry addition agents with the dry ground cement. When such a practice is followed, a master batch is prepared containing the addition a ents and some Portland cement or gypsum as a diluent in order that the very small quantity of the addition agent may be thoroughly distributed throughout the whole mass of ma- 5 terial.
In order that our invention may be better understood, we advance a theory of operation which our experimental evidence appears to bear out, but We do not desire to be bound thereby. 'With the exception of coal, which is frequently added to cement grinding mills in order to purge the mill or to act as a grinding a id, most cement addition agents seem to be substances which are very strongly adsorbed upon the surface of the individual cement particles. When these are added to cement, the surface energy requirements of the cement particles are satisfied by the adsorbed layer and no bonds exist to attract other particles. Hence, treated, dry cement does not form aggregates.
Apart from theory, we note as a fact that when water dispersions of cement addition agents are added to cement during the grinding operation, few aggregates are formed and the production of the mill rises, first, because the air separators then pick out the individual particles which are reduced to acceptable size, and, second, (perhaps because the addition agent is also adsorbed on the mill walls and grinding media), the mill does not become caked and grinding is much more effective. Surprising as it may seem, water may be added to cement clinker as the clinker is being ground without in any way injuriously affecting the cement or adding to the water content of the finished product.
This seems to be brought about because water itself is adsorbed upon the cement particle where it remains as a surface layer and, until the surface requirements of the cement are satisfied, no hydration of the cement takes place. Furthermore, the temperatures reached by the mill are sufilcient to vaporize off the adsorbed moisture and all added water is gradually released as the grinding progresses.
In order to give the cement its maximum opportunity to pick up and hold a surface layer of the addition agent, we add a wate dispersipn gilthe agent to the cement while it is being ground. The vapor pressure of our addition agents is quite low. They are not given off or released by the cement, but persist as surfacelayer, adsorbed substances which preventlr ig y agg regates from En ijig aHd later ''ii'rt their individual'ancTfi culiar effects upon the cement as it is used in concrete.
In practicing our invention, we first disperse the desired proportion of reagent to be used in water and then feed it to the cement clinker at some convenient point, usually as the clinker leaves the proportioning table or between the fuller or fiergul es mill nd the finisl mil}. It may, however, be added at other pointssuch as to the rejects of the air separator or to one of the ferent places, for example, one portion may be. added to the clinker as it is being fed from the, table and another portion added to the air sepa as we rator rejects. In general, we prefer not to use more than one part of water to 50 parts of cement or less than one part of water to 1,000 parfs of cement. Conditions, however, vary from mill to mill and we do not wish to be limited as to the 75 amount of water used, for sometimes we find it compartments of a ggrnpqb mill. At times we find 9 it advantageous to add the reagent at several dlfroom ation of h droxv-alk 1 amino and Fain sglfihonate f, 'gafi than 5m s of gin er a e as a 15 0 solution (this comadvantageous to use greater or lesser quantities than those above specified.
Some cement addition agents are soluble in the water used as a "carrier, other addition agents form so-called colloidal solutions or suspensions. This may be best demonstrated by specific examples. Tucker et al. use hydroxy-alkyl amines as strength producing reagents n concre e. havT odfidifitvvfiffsd caiied true solutions, 1. e., molecular dispersions, of these substances in waterarejdddtotlie "clement clinker in the mill, grindabili t y of the clinker is increased, resulting i'rTdncr ased production and" decreased power consumption per unit of cement produced. The grinding media are constantly kept clean and shiny; flaking of the cement is prevented; the limit of free grind is materially raised; the temperature gradient in the mill is greatly lowered; and the finished product is more uniform. Subsequently, when the ground, finished cement is used in concrete, the strength of the concrete is increased to a somewhat higher degree than when (as recommended by Tucker) the hydroxy alkyl amiggs are added to the gauge wa er. e spersing agents mentioned in the Tucker application (sole) form colloidal solutions in water, as also do those mentioned in the Mark application. When added to the clinker in the mill, water solutions of these act in the same manner as do the hydroxy alkyl amines, but in addition have the further functon of promoting a high degree of dry dispersion of the cement particles which results not only in cleaner grinding media, but also in a more emcient action of the air separators.
The effect of following our invention as set forth is much greater in commercial units than in laboratory mills. This is due to the fundamental difl'erence in the mode of operation of the two mills. Commercial grinding is a continuous process and gives poor opportunity for perfect distribution of dry addition agents due to the rapid rate of passage of the cement through the mill. The grinding in a laboratory ball mill, on the other hand, is discontinuous. The clinkers and agent are sealed in the mill until ground to proper size, thus much better distribution of dry material is effected. The more perfect distribution, however, due to following our invention, is shown even in laboratory mills by the increased grindability of the charge. Using a 20 mesh crushed clinker and an 8 to 1 steel ball charge in a steel mill the following surface areas, as measured on a Klein turbidimeter checked against a Bureau of Standards sample, were obtained after 2500 revolutions of the mill.
Surface areas, squarecentime ters per gram Blank With agent Li nin sulhonat 1 part to 2,000 parts of a 00 as a 136% solution 1745 1920 rox '-a k l amine 1 art to 7,000 parts of I asa A solution 1880 1935 ma niorms theh o u u I The drawing illustrates graphically a set of directly comparable grinds. The clinker used was manufactured in Pennsylvania from local materials. Each grind was made upon the same clinker in the same mill at the same speed of rotation.
watotwOSmONS, Y comma on PLASTlC The weight of the clinker and the weight of the balls were the same in each instance.
The best combination of grinding agents (number 6) increased the surface area of the cement 460 square centimeters per gram of cement above the blank reference grind made according to the usual practice.
It should be understood that such a set of curves is illustrative only. On some clinkers the efiect is greater, on others it is less than is here illustrated and each mill grinds in a somewhat different manner.
But, whatever the source of the clinker, Eastern, Central, Southwestern or Pacific, there is always a material increase in the output of the mill when our invention is followed and much less aggregated material is returned by the air separators.
In previous practice serious mechanical diificulties were encountered when cement addition agents were added to the clinker because these agents are utilized in extremely small amounts usually from .03 to .1 of 1%. Consequently, the proportioning and feeding of such exceedingly small quantities into the mill has been a major problem. By using a dilute dispersion of the agent a considerable volume exists which makes it practical to use many of the available types of liquid proportioning pumps to feed continuously the correct amount of agent to the mill, and since the feed may be continuous and not intermittent, distribution of the addition agent over the cement Examiner particles is very much better than any distribution which prior practice permits.
We believe ourselves to be the first to discover that cement addition agents, used for producing profound physical changes in the wet mix or for accelerating the high early strength, are made even more effective when ground with the cement clinker and, also, that these materials act as grinding aids, keep themills clean, and materially increase the lifiiit'of free grind. We have further discovered that, when the grinding agent is composed of the materials given in our preferred example, laitance after pouring the concrete is greatly reduced and the cement, technically, is non-bleeding.
Our discovery of the astonishing effectiveness of such cement addition agents as aids in grinding makes it now unnecessary to add deleterious substances such as rosin, tallow and coal to maintain the grinding rate of the mills.
What we claim therefore is:
l. The process of grinding cement which includes the step of adding a water dispersion of a lignin compound to the cement clinker in an amount insuflicient to cause the substantial hydration of the resulting cement and thereafter grinding the clinker.
2. A process in accordance with claim 1 in which the lignin compound is the hydroxy-alkyl amine salt of lignin sulphonic acid.
HENRY L. KENNEDY. JACOB G. MARK.