The Recovery of Potash as a By-product in the Blast Furnance Industry

Historical and technical developments of potassium resources. Davide Ciceri , David A.C. Manning , Antoine Allanore. Science of The Total Environment ...
0 downloads 0 Views 311KB Size
THE JOURNAL OF INDUSTRIAL A N D ENGINEERING CHEMISTRY

302

TABLEVIII-COMPARISONS OF NITRATESBETWEEN CHECKS AND APPLICAT I O N S OF NITRATEAT PLOWING TIME, SEASON 1921 Nos, P. p. ni. Lbs. of Dry Soil Date Check Plots Nitrate of Soda Plots 351 May 24.. 116’ 262 57 June 10................. G6 192 June 23.. 51 102 July 7 . . 45 94 July 21 58 August 5 . . 160 August 19... . . . . . . . . . . . . 49 92 47 100 September 1 1 Each figure is the average of five comparisons.

...............

............... ................ ................. .............. .............

an application of 400 lbs. per acre we have evidence that properly cultivated vineyards do not require nitrate of soda applications after plowing time. The check plot data are evidence that OUT present vineyard soils contain enough organic matter so that, wit,h good cultivation, bacterial activities will furnish nitrates in excess of the quantities used by the growing plants. The ratios of oxygen, carbon dioxide, organic matter, bacteria, fungi, plant food, etc., for the optimum development

Vol. 14, No. 4

of any variety of plant are not known. The synopsis a t the beginning of this paper gives merely indications of what the results secured may mean.

BIBLIOGRAPHY 1-W. A. Albrecht, Missouri Agr. Expt. Sta., Bull. 172. 2-P. L. Gainey, Soil Science, 3 (1917), 399. 3-P. L. Gainey and L.F. Metzler, J . A g r . Research, 11 (1917), 43. 4-C. A. Jensen, Bur. Plant Ind., Bull. 173. 5-C. B. Lipman, 35th Annual Meeting, Society Promotion Agricultural Science. 6-Lohnis and Green, Cenlr. Bakt. Parasitenk. 11 A b l . , 37 (1913). 7-T. L. Lyon and J. A. Bizzell, Cornell Univ. Agr. Expt. Sta., Memoir 1. 8-H. A. Noyes and S. D. Conner, J . Agr. Research, 16 (1919), 27, No. 2. 9-A. E. V. Richardson, J. Dept. Agr. Victoria, 11 (1913), 429. IO-T. Schloesing, Compt. rend., 77 (1873), 203. 11-A. L. Whiting and T. E. Richmond, Illinois Agr. Expt. Sta., Bull. 233. 12-A. L Whiting and W. R. Schoonover, Illinois Agr. Expt. Sta., Bull 226. 13-C. G Woodbury, H. A. Noyes and J. Oskamp, Indiana Agr Expt. Sta., Bull. 206.

The Recovery of Potash as a By-product in the Blast Furnace Industry’2z By William H. Ross and Albert R. Mer9 BUREAUOP SOILS, WASHINGTON, I).C.

The weighted average of the potash in the ores, coke, and limestone used in the blast furnace industry amounts to approximately 0.2 per cent for each material, which is less than one-thfrd as great as that found for the raw mix used in the cement industry. In the case of the ores the potash ranges from 0.05 per cent for Mesaba ores to ouer 2 per cent for certain foreign ores. As the consumption of high potash ores is relatiuely small as compared with lour potash ores, the weighted auerage of the potash in the ores consumed is less than the mean auerage found for different ore samples. On the basis of weighted averages the total potash in the ore, coke, and limestone used in blast furnaces amounts, respectively, to 7.2, 1.9, and 4.9 lbs. per ton of pig iron, or to a total of 14.0 lbs. The potash in the slag amounts to 8.4 lbs. which [eaues a balance for the potash volatilized of 5.6 ibs. per ton of pig iron. This amounts to a total for all plants of about 100,000 tons annually as compared with 87,000 tons for the cement industry. As there are twice as many blast furnaces as cement plants in the United States, the quantity of potash lost per indiuidual plant must therefore be less in the blast furnace industry than in the cement industry, but it is possible that the dust from some blast furnaces in which manganiferous or southern ores are used may be richer than the richest cement dust.

a previous publication4 an account is given of a survey that was made of the potash that escapes from the cement plants of the country. Representative samples of raw mix and of the corresponding ground clinker were collected from all but four of the cement plants then operating, By determining the potash content of these different samples and knowing the ratio of raw mix to ground clinker, it was possible to make a close estimate of the potash lost from each individual plant. The results obtained varied

I“ ‘

1 Presented before the Division of Fertilizer Chemistry a t the 62nd Meeting of the American Chemical Society, New York, N. Y., September 6 to 10, 1921. Published by permission of the Secretary of Agriculture. 3 Analyses by William Hazen, R. M. Jones, T. R . LeCompte, W. B. Pope and G. J. Hough. 4 William H. Ross,Albert R. Merz and C. R . Wagner, “The Recovery of Potash as a By-product in the Cement Industry,” U.S. Defil. Agr., Bull. 672 (1917), 1. f

from 0.35 to 5.14 lbs. of potash per bbl. of cement produced, with an average for all plants of almost 2 lbs. On the basis of an average production of 90,000,000 bbls., the total potash escaping from all the cement plants of this country was estimated to amount to about 87,000 tons annually. Since these estimates were made, installations have been placed in a number of plants for the recovery of this potash, and in every case the quantity that was found to escape from the kilns closely agreed with the estimates that had been made during the course of the survey. It is therefore logical to assume the reasonable correctness of the total estimate that was made for all the plants of the country. METHODOF INVESTIGATION On the completion of this investigation a corresponding survey was undertaken of the potash that escapes from the blast furnaces of the country. Through the cooperation of Mr. Frederick Crabtree of the Bureau of Mines a systematic collection was made of iron ore, coke, and limestone samples which represented the principal districts now producing these materials for use in blast furnaces. Each of the samples was analyzed for potash by a t least two chemists. From the data thus obtained, it was possible to calculate the quantity of potash that was charged into a blast furnace during any period if the source and consumption of the materials used in the charge were known. ThiP quantity less that which was lost in the slag gave the total potash that escapes from the furnace. Information on the averagk amount of potash lost in the slag was secured by analyzing a large number of samples collected from many of the leading blast furnaces of the country.

RESULTS The results obtained and summarized in Tables I and I1 show that the potash content of the numerow samples of ore, coke, and limestone that were analyzed give a weighted average about the same for each material, which amounts to approximately 0.2 per cent. In the case of the ores the

THE JOURNAL OF INDU8TRIAL AND ENGINEERING CHEMISTRY

April, 1922

potash ranged from a minimum of 0.05 per cent for some Lake Superior ores to over 2 per cent for certain foreign ores. As shown in Table I, samples from the Menominee and Marquette ranges contained between two and three times as much potash as those froin the Mesaba range. The potash in the ores from the same range agreed so closely that no material difference was found between the mean and weighted averages for any range. This close agreement prevailed both for Bessemer and non-Bessemer ores. L Southern and imported manganiferous ores are usually richer in potash than the Lake Superior ores. The imported manganiferous ores vary very greatly in potash content, some being as low as those from the Mesaba range, but the average for these ores is greater than for any of the domestic $oresexcepting those from the Georgia district. TABLEI-POTASHIN D O M E ~ T IACN D FOREIGN ORES

SOURCE OF ORE Mesaba range Gogebic range Menominee range Marquette range Cuyuna range Vermillion range Alabama New York Virginia New Mexico Georgia Miscellaneous domestic Foreign nonmanganiferous Foreign manganiferous

Production and Importation Ores in 19201 Gross Tons Analyzed 36,641,880 76 26 8,298,206 26 5,651,542 22 4,457,600 8 1,757,775 1,053,518 5,894,011 920,009 320,924 274,219 104,511

Total Samples Analyzed 126 46 36 31 S 9 8 3 4 5 3

Potash (KnO) Weighted Average Per cent

0.09 0.10 0.28

0.23

0.10

0.11 0.55

0.10

0.32

0.10 0.82

2.230.261

10

11

0.51

1,273,456

5

6

0.14

11

22

0.77

600,0002

WEIGHTEDAVERAGE FOR ALL ORES

.

0 19

Geol Siirv , “Mmeral Resources,” 1920 (1922) P a r t I 2 Estimated 1

MA?ERIAL I r o n ore Iimestone Coke

Consumption in 19201 Short Tons 77 800 000 ~O:OOO:OOO 42.000.000 . .

OF PO?.4SH

IN

Average Potash (K20) Content Per cent 0.19 0.20 0.24

TOTAL IN CHARGZ, , .... , , . . . . . . . . .

BLASTFURNACE INDUSTRY Total Potash (KzO) Pounds Potash Content (KzO).per Ton Short Tons of Plg Iron 147,820 7.2 40,000 1.Y 100,800 4.9

-

14.0

173,600

8.4

TOTAL LOSTB Y VOLATII,IZA?ION, . , , . . . , ,116,020

5.6

Production 1920 41,300,000 31,000,0002

0.56

.

1

2

-

. . . . . ,288,620

Pig iron Slag

__

__

Geol. Surv., “Mineral Resources,” 1920 (1922), Part I. Estimated.

In an investigation that was made some time ago by R. J. Wysorj of the Bethlehem Steel Co., it was reported that the potash in the ore, coke, and limestone used at that plant amounted, respectively, to 10.3, 5.8, and 6.3 lbs. per -ton of pig iron, or t o a total of 22.4 lbs. per ton. 5

Bull. A m . Inst. Mining Eng., 121 (1917), 1.

The results given in Table 11, which have been obtained for numerous other plants throughout the country, show corresponding values of 7.2, 1.9, and 4.9 lbs. per ton of pig iron, or a total of 14.0 lbs. The average potash in the slags, on the other hand, was found to be almost twice as great as reported by Wysor for the Bethlehem Steel Company. It therefore follows, as Wysor himself predicted, that the potash volatilized per ton of pig iron must be less for the average plant than was found for the Bethlehem plant. According to our results this amounts to only 5.6 lbs. per ton of iron as compared with Wysor’s 17.9 lbs. That the potash lost at the Bethlehem plant is about three times greater than the average for the blast furnaces of the country as a whole is further indicated by the potash found in the dust collected in the stoves and boilers of different plants. The average given by Wysor for dusts collected at the Bethlehem plant amounted to 9.9 per cent. The average value which we have found for the potash in sixtysix samples of dust collected a t various other plants amounts to only 3.3 per cent, or one-third of that for the Bethlehem plant. If our value of 5.6 lbs. be taken as the average quantity of potash volatilized per ton of pig iron and an average normal production of iron of 36,000,000 short tons be assumed, then the total potash volatilized from the blast furnaces of this country will amount to about 100,000 tons, as compared with the 87,000 tons found for cement plants. As there are twice as many blast furnace plants as cement plant4 in the United States the quantity of potash lost per individual plant must therefore be less in the blast furnace industry than in the cement industry.

POSSIBILITY OF RECOVERY

It happens, however, that the consumption of southern and manganiferous ores in this country is only about 10 per cent of the total, while the production from the Menominee and Marquette ranges is less than one-third as great as that from the Mesaba range. The weighted average of the potash in the ores consumed in this country is therefore less than the value obtained for the mean average of potash in different ore samples. According to the results reported in the Department of Agriculture bulletin already referred to, the potash in the raw mix of cement plants varied at the time of the survey from 0.2 to 1.16 per cent, with a weighted average for all the plants of 0.67 per cent. The averagepercentage of potash in the raw mix used in cement manufacture is therefore over three times as great as that occurring in the raw materials consumed in the blast furnace industry. TABLE11-ESTIMATED LOSS

303

When the survey was made of the cement industry it was recognized that the dust from some plants was probably too low grade ever to be used as a source of potash even if no cost were involved in collecting the dust. The richness of the dust in other plants indicated, on the other hand, that potash could be profitably recovered from this source, particularly if some inexpensive means were installed along the lines of the process now in operation at the Santa Crux plant whereby a mechanical separation of the potash and the dust is brought about during the process of collection. In the blast furnace industry the greater part of the volatilized dust is now lost in the primary washers. The possible recovery of dust in this industry is therefore dependent on the substitution of a dry system of purifying the gases in place of the present wet system. Judging from results that have already been obtained in several plants with two dry cleaning systems of entirely different design it is possible that a substitution of a dry system for the present met system may ultimately be made in many blast furnace plants. Owing to the low potash content of the materials used in some plants it will no doubt be found that the dust from these plants if collected will be too low grade t o be used as a source of potash unless it is found to contain other products of value which will contribute to the cost of concentrating the potash. The situation with respect to plants operating on manganiferous, southern, or Menominee ores should be entirely different. Owing to the richness of these ores it is possible that the dust from these plants will be found to be richer than the richest cement dust. It may further be emphasized that if dry systems are eventually installed for purifying blast furnace gases the collected dust, although possibly not extensively used as a source of potash under normal conditions, might nevertheless serve as an important potential source of potash which would be immediately available in caSe of a future emergency.