SELENIUM IN MEXICO HORACE 0. BYERS Bureau of Chemistry and Soils, Washington, D.C.
D
URING February and March, 1937, the writer, accompanied by John T. Miller and a representative of the Mexican government, explored several areas in Mexico which were suspected of being seleniferous. I n the valley of the Guanajuato River, State of Guanajuato, there has existed for more than 200 years an obscure disease known as soliman disease because of its presumed source in the mines in and about the city of Guanajuato. Soliman is
even though mercury had not been found in either soil or plants from the area. These facts were reported orally to A. B. Clawson, of the Bureau of Animal Industry, by Jose Figueroa, of the Instituto Biotecnico of the Department of Agriculture of Mexico, and relayed to the writer through T. D. Rice of the Soil Survey. Opportunity was then sought to look into the matter because the symptoms as given in detail by Roca coincided exactly with those described by Franke, Rice, Johnson, and Schoening (2) for the “alkali disease’’ of South Dakota and elsewhere which has been shown to be due to the presence of selenium in certain soils and plants. In making the field examination, attention was directed specifically t o determining (a) if the cause of the disease of both men and animals could be definitely assigned to selenium, (b) whether the area affected is limited to the flood plain, and (c) if the source of the selenium is the mines or is inherent in the soil itself. No attempt was made to determine the extent of the area affected or the extent of the injuries to either men or animals. These and related questions appear to be the problem of the local authorities, Samples were secured from the following sources: ores presumably representative of the area, slimes from the waste dumps, water from the river below the slime heaps, soil subject to silt deposit from the river a t flood levels, similar soils not subject to overflow, plants of similar types on both soil &rem, feed of animals affected by the disease, and food materials consumed by the people of the area. A sample of milk was also secured from a cow which was characteristically diseased.
TABLE I. SELENIUM CONTENT OF ORES,GANGUE,AND EXTRACTED PRODUCTS Lab.
No.
Field No.
B20879
12 12A
Sirena mine above Guanajuato
Slime heaps
12c __ 12D
320880
12E 12F 120 12R 13
B20881
14
B20882
15
B20883
16
B20884
17
B20885
18
B20886
19
B20886A B20887
19
20
Location
Type of Material
2.0 2.0
Mine waste
2.0 2.0 2.0 1.0 30.0 13.0 2.0
Moist silt slime a t base of J u m Moiat clayey sfime a t foot of dump Silty slime 8 ft. from bottom Laminated male a t t o of dump Fresg slimes at top of dump Fresh scale on stream bank Ouanajuato river 0.25 Water from stream mile below slime (sol. Se) dump Same Silt settled from water Marfil (abandoned Adobe wall of town town)
the Spanish name for mercury chloride, which wm believed to come from the patio process for extracting silver and gold from the ores. The slimes left after extraction were deposited in the narrow canyon and in flood times deposited over the flood plain of the stream. For several reasons the incidence of the disease was a s s o c i a t e d with the mines. Among them was the fact that food for both animals and man appeared to be toxic only when produced on land flooded a t intervals, and therefore silted or irrigated from the water of the stream. Also the disease symptoms were s o m e w h a t similar to those observed in mercurial poisoning. An unpublished report by Juan Roca in 1930 reported mercury as the probable source of the trouble
Selenium P. p . m. 4.0 35.0 2.0 2.5 65.0 2.5
4.0 5.0 5.0 6.0
Analytical Data
3.5
Table I gives the results of an examination of ore samples obtained from the Sirena mine and the waste dump of the mine. All of them were uncrushed rock. No attempt was made to identify the minerals represented. According to sev-
4.0 0.2 8.0 64.0
CONTENT OF SOILSAMPLESFROM TABLE 11. SELENIUM VALLEY NEAR IRAPUATO Lab.
No.
Field No.
Location
THFI
GUANAJUATO
Type of Material
Depth
In. B20861 B20866 B20868 B20872 B20877 B20896
2
3 4 7 10 39
B20897 B20901
39 39 R20903 __....39 B20904 26’ B20905 27‘ B20906 27’ B20910 28 B20911 29 B20916 31 32 B20917 B20920 B20923 B20926 B20927 B20929
33 34 35 35 36
B20931
37
B20945 B20860 B20946 B20948
41 1 42 43
Sonaja ranch: 4 mi. south of Irapuato 80 rods south of sample 2 1 6 mi southeast of ranch house 2’mi. sbutheast of ranch house 0.25 mi. northwest of ranch house 1 mi. south of Irapuato sewage disposal in excavation Same Same Same Garrrida ranch: 4 mi. north of Irapuato 200 yd. north of ranch house Same 260 yd. north of ranch house In depression near 27 ft. Same
Same 1 mi. north of Irapusto near roadside Jesus Velassuez ranch, 1 mi. northeast of Irapuato 0.5 mi. east of Irapuato 17 west of Queretaro 1 m.east of Salarnanca 12 mi. east of Queretaro
I+.
1200
Black clay Same
0-8
Dark gray cla loam
&
Recent river osit Dark gray claytoam
Black clay Grayish brown clay Black clay Dark red clay Soil and manure in corral Fine sandy loam Same Sandy loam Black clay Old adobe wall
Selenium P. p . m.
0-8
1.0 0.8 4.0 3.5 0.3
0-10 10-25 65-97 144-216
1.5 0.3 0.2 0.1
0-8 0-8 0-8
...
0-8
8-12 0-8 0-8
...
20.0 4.0 4.0 4.0 4.0 4.0
Dark gray clay loam Dark clay loam BIack clay Gray clay loam Same
0-8 0-8 0-6 6-12
0-8
4.0 1.5 4.0 6.0 6.0
Black clay loam
0-8
1.0
Dark gray clay loam Black clay Same Black silt loam Heavy black clay
0-8 0-8 4-12
6.0 0.4
0-8
0-8
0.3 0.4 0.4
OCTOBER,1937
INDUSTRIAL AND ENGINEERING CHEMISTRY
1201
era1 independent sources, these ores are representative of the entire region in question. The data show that the seven assorted samples of ore have a Lab. Field mean selenium content of 16 parts per million. This is the No. No. Looation T y p of Material Selenium material which, when crushed and silver-extracted (formerly P. I). m. by the patio process) constitutes the slimes washed into the 820862 2A Sonsjs ranch Common mustard 1 pits near the mouth of the canyon. T h m enormous dumps 820803 2 8 Coaklebur 2 820864 2C Immature wheat ha& are the real source of the stream. The unmilled mine waste and ieaves l 820865 2D Chick-pea 4 is rejected material low in silver; it was examined because it B20807 3A Alfalfa 3 is high in marcasite and other sulfide and was expected to be B20809 4A Ground oherry 8 B20871 0A Ground sherry 35 high in selenium. The mean content of the samples was 820873 7 A LBmb’s-aUartBrs 2 820874 7B ImmatUIs whest heads only 7 p. p. m. The slime dumps are fed many thousand tons and iesvea 10 of fresh material daily and represent the accumulation of many Milkweed 20 years; a t irregular intervals they are tremendously eroded by floods, carried downstream, and distributed over the flood plain of the Guanajuato both above and below the town of Irapuato. The mean selenium content of this material, as shown by the six samples, is 4.6 p. p. m. Unidentified legume I8 The stream flows rapidly from the base of the dumps for Johosoa 12 MilkwwrBS8 3 some distance below the town and carries so much suspended Wild lettuce 4 Johnaan wasa 0 material that it has a milky appearance. The water is used 120 Common mustard to a considerable extent for irrigation. The analysis shows AlfSUa 30 Lsmb’s-quaitera 19 that it contaias a definite selenium content and that its deposAlfalfs (from manger) 18 ited silt is markedly seleniferous. The most remarkable Milk from c(tw eSting 38A 0.8 sample is that taken from the wall of an abandoned town some 3 miles below the slime dumps. The wall and houses were said to be built of “adobe” bricks made from the slime and organic material, probably manure. The wall is very old and its high selenium content bears out the tradition that the soliman disease in the valley below has, existed ever since these mines began to be worked, probably more than 200 years. Table I1 gives the resnlts of the examination of soil samples collectedin the valley. Included are soilsfrom the flood plain ranch is of long standing. Samples 33 and 36 are from fields of the stream which are subject to deposition of silt from the on the Arandas ranch wEch are infrequently flooded and not mines, and a number of samples from places whioh are not irrigated. Sample 35 is from a field pointed out by an Indian flooded. Included also are certain samples of soils of the as the spot where corn grew that wa8 responsible for hia own same character from other areas not adjacent to Irapuato but case of alkali disease. Sample 37 is from a field on the Jesus derived from similar parent material, except alluvial deposits. Velasquez ranch. Alfalfa grown there was reported as the The data given in Table I1 must be considered in the light cause of the lass of 125 head of cattle in a night’s feeding. of the following facts: Samples 2 and 3 are from an area Obviously the source of serious quantities of selenium in the flooded from the river only infrequently and irrigated with soils about Irapuato is the silty material washed down from water from a deep well presumably not seleuiferous. Samples 4 and 7 are from land flooded whenever the river overflows its the mines about Guanaiuato. normal banks. Both samples are Table 111 gives the results of the more silty than 2 and 3. Sample examination of samples of vegetation 10 was taken on a small hi11 presurncollected in the area about Irapuato. ably never flooded. I t contains the Results obtained from the samples normal quantity of sclenium for soils collected on nonseleniferous spots are not toxic and corresponds to samples omitted. They were essentially free 1,41,42, and 43. Sample 39 is part from selenium. Included are nine of a profile taken from the banks of samples bought a t random in the an excavation for a sewage disposal Irapuato public market. plant which is being erected. The All the vegetation grown on soil area is subject to flooding approxiknown to be seleniferous contains mately three times in 10 years, and selenium. That grown upon the the soil is probably an alluvial deSonajaranch is definitely lower than posit. Only the recent surface conon the other spots examined, probtains selenium in excess of the normal ably because some of the irrigation traces found generally in soils. water comes from a deep well preAll the samples taken on the sumably free from selenium. In the G a r r i d a r a n c h contain abnormal othersamptes, irrigation water comes quantities of selenium and all are from the river which contains disfrom the river flood plain. Two solvedselenium (Table I, ample 19). of them are of special interest: SamIn severalareasin the Unitedstatesit ple 26’ from the corral indicates that has been shown that irrigation tends the manure used for fertilizer is exto diminish the effects of selenium on ceptionally rich in selenium, and a m plants (1) but not when theirrigation ple 31 from the adobe wall shows w a t e r contains selenium. Sample that the seleniferous condition of the 35A is from the spot represented by FIQURE1. BADLYSELENIZED Cow
1202
INDUSTRIAL AND ENGINEERING CHEMISTRY
VOL. 29, NO. 10
soil 35 of Table I1 and was the only vegetation growing there when the examination was made. Sample 38A is alfalfa which was being eaten by a cow definitely suffering from the selenium poisoning (Figure 1). The milk from this cow contained 0.6 p. p. m. of selenium, which made up about 6 p. p. m. of the milk solids. Incidentally, this cow, as well as two bulls also suffering from the same trouble, was a newcomer into the valley. Some of the members of the herd of some fifty cattle showed evidence of the disease to a much smaller extent. The samples of vegetables sold in the city market are significant. Of the nine samples analyzed, all but two are definitely seleniferous, and these may not have been grown in the valley.
When we consider that more than 4 parts per million of selenium in a dry diet are definitely injurious (3)and that the people of Irapuato live largely on their local produce, it is clear that this location is an excellent one to study the effect of selenium in foods upon humans. Since selenium may be present in all the vegetables, milk, and meat consumed ( I ) , the citizen of Irapuato is likely to be affected by selenium from even his prenatal days. To what extent the evident ill health of the inhabitants is due to selenium in their food is problematical but is worthy of study,
Conclusions
In making the field examination here reported the Mexican Government was represented by Jose Figueroa; without his local knowledge and energetic assistance the collection of m% terials would have been almost impossible. The laboratory examinations were made by &. T. Williams and H. W. Lakin of the Soil Chemistry and Physics Research Division of the Bureau of Chemistry and Soils.
I n the area of seleniferous soils about Irapuato the toxic agent is evidently deposited in the soils from the waste product of the mines about Guanajuato. No similar situation exists, so far as is known, in the United States. The condition in Irapuato seems less serious than would be the case if selenium-loving plants, capable of enriching the soil with available selenium, were present. In this area irrigation as practiced is not a remedial measure ( I ) , partially because the region is semihumid and therefore does not require continuous irrigation and partially because of the presence of selenium in the water. This point may be important in connection with the use of drainage waters from seleniferous areas for irrigation in other places.
Acknowledgment
Literature Cited (1) Byers, H. G., U.S. Dept. Agr., Tech. Bull. 530 (1936). (2) Franke, K.W., Rice, T. D., Johnson, A. G., and Schoening, H. W., Ibid., Circ. 320 (1934). (2) Munsell, H. E., Ibid., Tech. BUZZ. 534 (1936). REcEIVBD
J d Y 21, 1937. 8
Investigating Time-Temperature Effects upon Properties
of Reactants W. M. BREITMANN 20, Riga Prospect,Ap. 38, Leningrad, U. S. S. R.
T
HE control of the chemical conversion of matter is the province of chemical kinetics, which studies in a dynamic aspect reactions that “static” chemistry investigated in detail over a century ago. Modern chemical industry urgently needs a technic that will make it possible to predict the effect of varying time and temperature conditions in a reaction upon the properties of the product. It is therefore natural to seek an empirical generalization that will enable us to control the chemical conversion of a substance by studying its physical and chemical properties during the course of the reaction. Certain regularities discovered by chemical kinetics can be used for this purpose. When the composition of the starting mixture remains constant, the properties of the products obtained are governed by the time and temperature conditions prevailing during the course of the reaction. Because instantaneous heating and, subsequently, cooling of a reaction mixture is always impossible, we may consider that every large-scale chemical process is proceeding a t a variable temperature. This is particularly
true when the reaction evolves or takes up heat. I n many cases the inevitable temperature variations are such as to make it difficult or even impossible to obtain products with the required properties. The present work is an attempt to find a method of investigation that will relate the physicochemical properties of the product to time and temperature variations during the reaction.
Development af Equations Consider the equation of polymolecular reaction, and for the sake of simplicity let us assume that the amounts of different constituents present are equivalent:
where z = amount of substance converted t time a = original amount of substance IC p: reaction constant n order of reaction The reaction constant depends, as is known, on temperature: IC = Mf(T)
where M = a coefficient of proportionality T absolute temperature =1
(2)
