RESATEON$HEP BE?WEEN GXYGEN REFUSE?! RATES, A5 #AifiLSfiEé} WITH THE FLATIWM MICRGELECTRGDE, AND PLANT GROWTH 'f’msis For m Mm 0? Ph. D. MlCHéGAN STATE UNWERSWY mm Méfiéoe' Van Daren, Ir. 1958 This is to certify that the thesis entitled Relationship Between Oxxgen Diffusion Rates, as Measured with the Platinum Microelectrode, and Plant Growth presented by David M. VanDoren, Jr. has been accepted towards fulfillment of the requirements for Jill.— degree ianiLSLiane Major professor Date May 20, 195;. 0-169 -vcm — .1 "‘1' w o '5‘: ”h find“..— i i L RELATIONSHIP BETWEEN OXYGEN DIFFUSION RATES, AS MEASURED WITH THE PLATINUM MICROELECTRODE, AND PLANT GROWTH By David Miller Van Doren, Jr. .A THESIS Submitted to the School for.Advanced Graduate Studies of Michigan State University of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Soil Science 1958 ACKNOWLEDGMENT The author wishes to sincerely thank Dr. A. E. Erickson for his guidance and assistance given throughout this research project. Dr. Erickson gave freely of his knowledge and time and provided the necessary equipment to complete this investigation. Appreciation is also ex- tended to the author's wife, Janet, for her aid in the preparation of this paper. RELATIONSHIP BETWEEN OXYGEN DIFFUSION RATES, AS MEASURED WITH THE PLATINUM MICROEIECTRODE, AND PLANT GROWTH By David Miller Van Doren, Jr. AN ABSTRACT Submitted to the School for Advanced Graduate Studies of Michigan State University of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Soil Science Year 1958 I AU." ilk T g 7317‘ Ale 1 a G c Cw WNW hi.» A E i I 6..» {in '1"- a. .ABSTRACT The purpose of this dissertation is to quantitatively evaluate the factors that influence the measurement of oxygen diffusion rates with the platinum microelectrode, and to evaluate the relationship between these rate measurements and plant growth. .Among the factors that influence the measurement are the length, degree of wetting, and.physical and chemical history of the platinum wire, temperature, pH, and electrical resistance of the soil and interference from reducible substances other than oxygen. The limitations imposed on this measurement of oxygen diffusion rates by these factors is discussed. Tomatoes, sugar beets and corn were used as indicator crOps of the relationship between plant growth and oxygen diffusion rate measurement. .An experiment performed in the greenhouse showed that short periods of oxygen stress may cause a considerable reduction in tomato vegetative growth, if such an oxygen stress occurs early in the tomato growth period. Tomato fruit was harvested prematurely and no such results are evident, In an.experiment conducted in the field tomato plants were subjected to varying degrees of oxygen stress over most of the growing period. Those plants grown in plots in which the diffusion rates were below a certain "critical" rate in the early portion of the tomato growth.period in general produced less vegetation and fruit than tomatoes grown in plots having diffusion rates greater than the "critical" rate for tomatoes. iv Field oxygen diffusion rate measurements were made in sugar beet and corn plots at the Ferden farm in 1955, 1956, and 1957. These measurements show that sugar beets grown in plots in which the diffusion rates were below the "critical" rate for sugar beets over an extended period of time during the first half of the growth period did not yield as well as beets grown in plots having diffusion rates averaging in excess of this "critical" oxygen diffusion rate. Corn did not follow the same growth.pattern with respect to oxygen diffusion rates as sugar beets. The possibility of relating the rate of change of oxygen diffusion rates following rainfall or irrigation to eventual crop yield is presented. Existing data related to plant growth and "critical" oxygen diffusion rates are summarized. 4- .7-‘ a-‘b t s p H TABLE OF CONTENTS mTROwcTIONC0.0IIOUCOOCCCOOOOODOOCIOOICO...-OCOIOOIOCUCOODOCCOIO I. FACTORS AFFECTING THE MEASUREMENT OF OXYGEN DIFFUSION CURRENT WITH THE PLATINUM MICROMTRODE.................. A. Physical Factors.................................... 1. Dimensions of the platinum wire............... 2. Exposure of copper............................ 3. Variability among individual platinum micro- electrodes................. .................... )4. Wetting of the platinum nfi.croelectrode........ 5. Temperature................................... B. Chemical FactorSIOIOOOOIOOOOOIOIOIOO0.00.0000...I... l. Polarisation of the reference cell............ 2. Poisoning of the platinum microelectrodes..... 3. pHIOOOII.0..IO...OI...-OIOOOOOOIOOCOOIOOOOIOII )4. Electrical resistance of the soil solution.... 5. Interference from other reducible substances.. II. RELATIONSHIP BETWEEN OXYGEN DIFFUSION MEASUREMENTS WITH THE PLATINUM MICROEECTRODE AM) PLANT GROWTH.............. A. Field mmg variationSIOIOOCOOOOO.OOOOOOOOIOOIOOO B. Evaluation of Critical Portions of the Growth Period 0f Tomboes.OOOCIOOOOOOOOOOOIOOCIOOOOO0.0.0.0000...I l. Grewhouse expermmtOOOOCOOOOOIOIO0.0.0.0.... 2. Field mermmtOOOCOIOOOIOIOOIOOOIOOIOIOOOCOO C. Evaluation of the Effect of Soil Aeration on Crop YieldOOOCOOOCCCOCOOOCCOOO0.0'DOOOCOOOOOIOOIOOCOIQOOO D. Use of Rate of Change of Oxygen Diffusion Rates Following Rainfall or Irrigation.................... E. Discussion.0‘....ICOIOCOOI‘ICOOOCOOIOOOOIOOOOOIOICOO WIOCCDOIUOOOIIOOOIIO...0IIOCQCOI...IOIOOCOOOOOOCIOOOOCIOOOI. BIEIOWH‘IOOOCCIOOOIOOOOOOOI0.0...I0.0COIOOIOOIIOOOOOOCOIOOIOO. 30 3O 37 55 58 63 ., . q \ - a n y A A I s 5 A a A 0 ~ - a s ' n n \ v) n . I h - ~ ! Q P 'c 0 t II n " O o v - r - . . I. r I- ” a a o a. a '1 . . A ~ '- r a 9 ~ 4 . - A . a a - ~ w a A A w o a r 5 L r ~ ~ - I - 5 3 .. . 4 a 4 a - . ,— . . o .. . a 9 n r A ~ I h ‘ ‘ I I ~ I O “ R Y i N l n r ' I Q ' c - s . u . . . a , . , .. - q . ‘ ‘ . a - r u - . . n a q; n r . , u .- .- ‘ u .- II a o u a o . 1 a . 1 a I: ~ . r- 7 u q - n a s > 3 m a e . r 1 a I! r r .- ~ - q o- a v I a o v ‘ I r A a :- . - e a - " a I A ‘ e o \ P a a ' - c A A u 0' ‘I O l . P Q H I I I D u l I a Q.‘....:Q'QCIID'I'U'...‘\.."D"..§.r.".-."”I - n~as I ‘ 1. I l C ‘V ‘ LL . .v’nnn.—..A’-- 3. . an. 1'n-g‘nu‘Iqfl"QA'IQ‘I‘A~0050‘! - Na. k. . C‘OOU”.‘O('EQ.‘IIIOAI‘I"‘.§I"I’J‘IOOCOO‘ 4‘ ..~-,.r‘aan~nrna errl' ‘ I.‘..‘I...’.III...'.RI.....-fi’ | S?! l. c . a”: “wit ~t.nannpnals--oscanoo’~-’-~vow-»n«n\’~r-«o.- .. c.~aun~oa~oaenv«v'o~ u u. "a ngh :Iu'.n.qhn« 0.1!..ngr.‘cvr.n.cao.:A-qu¢1.9 ~ “MD "5‘ ‘Ofl'. ‘..“’V.‘~“'n"f'.i‘..9.'F'I..-.’.".’.I’ IOIDI..". . t.‘ .— l 0Q"!I”!l.-DO'.IIIOOI'OBQ.‘Allrrl.tl[.n‘t04l0“’-O‘.OI.- L‘s-«g “he ‘vm INTRODUCTION The material presented in this thesis is a continuation and elaboration of the work initiated by Lemon (lo) and Lemon and Erickson (11,12) and continued by'Archibald (1), Cline (3), Jackson (7), van Doren (16), and others under the direction of Dr. A. E. Erickson. Discussions of the general relationships between soil aeration and plant growth by Russell (15) and the use of the platinum microelectrode as a method of effectively measuring soil aeration conditions in_§itu (10,11,12,16) are available. It is the purpose of this study to: (a) Quantitatively evaluate the factors affecting the measurement of oxygen diffusion with the platinum microelectrode. In principle this method for measuring oxygen diffusion is a single point amperometric determination for oxygen. That is, under the proper conditions (discussed in Section I) oxygen will be reduced to two 0-2 ions at the surface of a small platinum wire by applying a given potential between this wire and a suitable reference electrode (8). The current produced will be directly proportional to the amount of oxygen reduced. After about 5 minutes the platinum wire becomes concentration polarized with respect to oxygen. In other words, after this period of time has elapsed a minimum limiting oxygen diffusion rate is approached and the system approaches a steady state (10). The oxygen diffusion rate is determined by measuring the current produced by this reaction in the steady state (equation (g) p. 6 ). (b) Relate the growth of greenhouse and field crops to soil aeration as characterized by this measurement. Oxygen is essential for the respiration process in plant roots (13). Without a sufficient rate of supply of omgen in the soil the roots and eventually the entire plant will suffer. The usual overall effect is a stunting of growth and reduction in yield. By sufficient rate of supply of oxygen is meant a flux of oxygen great enough to maintain the concentration of oxygen at the root surface above a "critical" level (19). Below this ”critical" ongen concentration level, the respiration rate of the roots decreases as the rate of supply of oxygen becomes more limiting and a decrease in plant growth results. Meyer and Anderson (13) present a general discussion of the factors affecting respiration. Among these are rate of oxygen supply, as discussed above, and temperature. With increasing temperature the respiration rate and the need for oxygen increase. On days when the conditions for transpiration are optimum the uptake of water from the soil, which is partly dependent on energy derived from root respiration, is relatively high. Meteorological conditions that influence transpira- tion, such as incident radiant energy, wind velocity, temperature and relative humidity, are thereby indirectly responsible for a greater or lesser need for oxygen by plant roots. There are certain stages in the plant growth period when respiration rates are higher than at other times and/or when lack of sufficient oxygen my cause more damage to \ .) ) )) )))))I)‘ 1L1)“; (3))le rhlu a)! )p) II)).) x the plant than at other times. Therefore, in establishing a "critical" rate of omgen supply for a given plant species a great many seasonal and daily fluctuating variables must be taken into consideration. To further complicate the problem, oxygen diffusion rates in a given soil system differ from day to day and from hour to hour during a given day. The direct causes of such changes are changes in soil temperature and oxygen concentration in the soil atmosphere. Temperature effects will be discussed in Section I-A. The activity of oxygen in the soil solution is directly proportional to the concentration of oxygen in the soil atmosphere at constant temperature which, in turn, is greatly affected by biological activity. According to equation (1), page 6 , the diffusion rate is directly related to the activity of oxygen in the soil solution. The daily fluctuations of the oxygen requirement and rate of supply make the correlation of plant growth and oxygen diffusion rates very difficult. At any time on a given day the rate of supply might be low when measured, while the oxygen requirement of the plant is also low. In such a situation the rate of oxygen supply would not affect the plant nearly so much as an identical rate of supply when the omgen requirement was high. Since it is possible to get many combinations of rate of supply of oxygen and oxygen requirement on a given day and from day to day, the measured rate of supply may or may not be correlated with plant growth, depending on the actual oxygen requirement of the plant at the time of the measurement. It is postulated that in order to make valid experimental compari- sons of effects of a measured rate of oxygen supply on plant growth, the rate of oxygen supply measurements should be made at the same time of day on each day that such measurements are taken. Furthermore, not only a measure of the rate of oxygen supply is needed, but also such meteorological data as are related to the oxygen requirement of plant roots must be obtained. From such measurements under controlled conditions in the greenhouse, the "critical" rate of oxygen supply for various crop plants can be obtained as a function of the factors influenc- ing the oxygen requirement of the plant roots. With this information, field measurements of the rate of oxygen supply, along with other pertinent information, can be used to estimate the effects of oxygen supply on field crop yields.. It is virtually impossible to obtain all oxygen diffusion measure— ments from several locations simultaneously. However, it should be possible to obtain temperature and oxygen concentration of the soil atmosphere within a short period of time for these locations. Then if these same factors are again determined at the time the oxygen diffusion measurements are actually made (all on the same day), equation (k) (p. 7) and the known temperature coefficient of the oxygen diffusion current can be used to calculate the oxygen diffusion rate at the desired time of day. Unfortunately, the pieces to this puzzle have been put together slowly, and the necessary data for this sort of analysis are not all available. "Critical" rates of oxygen supply as herein specified are not available. Several investigators have obtained values of oxygen diffusion rates that have been critical to the particular plant studied under the particular circumstances of their greenhouse experiments. However, all of the necessary data were not obtained to make accurate comparisons of oxygen diffusion measurements and their effect on plant In growth in the greenhouse with oxygen diffusion measurements and their E effect on plant growth in the field. The available data are presented in the Discussion, p. 61. .~. gum n:;4:+ ~' ‘ ‘ g T»- ‘$". and chem A. Ph‘rsic The : measuremer the platin Rating of factors may factors 5115 1. Dim The eq.~ airface Cleve I. FACTORS AFFEXJTING THE MEASUREMENT OF OXYGEN DIFFUSION WITH THE PLATINUM MICROELECTRODE These factors may be grouped into two major categories--physical and chemical . A . Physical Factors The individual physical factors affecting the oxygen diffusion measurement with the platinum microelectrode include the dimensions of the platinum wire, exposure of copper, physical history of the platinum, wetting of the wire, and temperature. In the future other physical factors may be recognized, but at present the preceding are those factors suspected of influencing the measurement. 1. Dimensions of the platinum wire: The equation representing steady state diffusion to a cylindrical surface developed by Wiegand and Lemon (19) is: f _ 21¢. De (op-or) 60 sec/min (y w A (in re - ln'R) where fw a weight of ongen diffusin to a unit surface of platinum wire in unit time (gm cm" ’ ‘1) L - length of platinum wire (cm) A surface area of platinum wire (cmz) - 211M. De - diffusion coefficient of omgen in the solid—liquid phase (cm2 sec-1) Cp activity of oxygen in the liquid phase at a gas-liquid phase boundary (gm cm‘3) Cr .- activity of oxygen at the surface of the platinum I: 0 when t >0 (t - time) radius of platinum wire (em) -= distance from the center of the platinum wire to the point in the soil solution where de/dre - O R 1'e In terms of the oxygen diffusion current, i: f _ i M 60 secszn (2) W nFA '- where i a diffusion current in microamperes M a molecular weight of oxygen = 32 gm/mole n . number of gram equivalents per mole of oxygen = h F = Faraday constant = 96,500 coulombs/gram equivalent A = area of the platinum wire (cm?) Combining equations (1) and (g): i M 60 seczmin _ 21mg op 60 sec/min (2) n 'A"'?E“—'—‘Tre — In R Simplifying: i " (anti) (Db—1%) L (1‘) Since diffusion current is directly proportional to the length of the platinum wire, the importance of maintaining microelectrodes at the same known length is obvious. 2. Exposure of copper: The reduction potentials of copper are -O.20 or —O.h8 volt depend- ing upon the oxidation state (Table I). These potentials lie between zero and the potentials used for reducing oxygen at the platinum surface. If copper oxide (CuO) on the copper wire used in making the platinum microelectrodes (16,17) were exposed to the soil solution, copper (+2) would be reduced to copper (O). This could occur if a small crack or break through the insulation were present from the body of the solution to the copper wire. Solution would diffuse into this break, wetting the copper surface and providing electrical contact between the copper and the reference cell by way of the body of the solution. The reaction at the copper surface so exposed would add to the overall reaction with the microelectrode (See Figure 1, curve 2). Subtracting ordinates of .. ,m 2 : _. TABLEI Reduction Half Wave Potentials of Inorganic Substances (Kolthoff and Lingane) Substance Supporting Electrolyte E-Q-vs. S.C.E. Notes Cobalt 60+: in 1 N muon + 1 N NH: Copper 1 N NH40H + l N NH4Cl Iron Fe+3 in K01, H01, etc. Fe+3 in 1 M menace3 Manganese Mn“ in 0.2 M Tartrate +2 M NaOH Molybdenum Mo+4 in dilute HNo3 + lactic and oxalic acids -O.3 volt Co+3 -> Co+2 -o.2o volt Cu+2 -> ou+1 -0.)48 volt Cu+1 -> Cu obtained at zero applied emf. —o.uh volt Fe+3 —> Fe+2 -o.u volt Mn+2 —> Mn+3 (anodic) -O.35 volt products not -0.5 volt known Reduction Half Wave Potentials of Organic Compounds (Kolthoff and Lingane) Substance Ei- vs. S.C.E. In general, E;- is more negative than —l.0 volt Fumaric Acid Maleic Acid Other Unsaturated Acids Nitrobenzene and Nitrophenols Organic Peroxides -0.§8 volt at pH 0.0 -0.71 volt at pH 2.0 -O.58 volt at pH 0.0 -0.71 volt at pH 2.0 Between —O.2 and -0.6 volt . at pH 0.0 Between -0.07 and-0.67 volt in the pH range of 1-9. Between -O.2 and -0.8 volt at all soil pH values. 9 Current Relativ U‘l Platinum and copper wires expo3ed. Curve 1: Platinum wire exposed. . A A A A FIGURE l. \n 1 3 3 L; Time in Minutes Current vs. time curves for platinum microelectrodes in C.l% KCl saturated 500 micron glass beads. lO curve 1 from curve 2, Figure 1 yields the portion of the observed current due to the reaction at the copper surface. Microelectrodes with only minute amounts of copper exposed may cause large errors in the measurement of oxygen diffusion. 3. variability among individual platinum microelectrodes: One of the difficulties with polarographic determinations is the effect of the history of the electrode on its reaction characteristics. When the microelectrodes are used in soil materials and not left in the medium for long periods of time these effects are mostly physical in nature. The microelectrodes become scratched and rescratched upon repeated insertions into the soil, and the dimensions of the platinum wire may become changed due to repeated bending of the wire. When the microelectrodes are left in the soil for long periods of time they are subject to "poisoning." This is a chemical phenomenon and will be discussed later. A standard medium was devised in which the 5 minute diffusion currents of all microelectrodes used in the field experiments in 1957 were obtained periodically during the summer of 1957. The object was to determine the change in reaction characteristics of the microelectrodes with continued use, if any, and to eliminate those microelectrodes that were not functioning properly. The medium was a 3 percent suspension of natural sodium saturated wyoming bentonite with 0.1 N sodium chloride solution as the supporting electrolyte. The purpose of the clay was to reduce the possibility of convection in the medium. Convection would increase the rate of movement of oxygen to the surface of the platinum wire, rendering the results nonreproducible. This standard medium is inexpensive and easy to reproduce. The microelectrodes were returned to the laboratory from the field every 2 weeks. Each microelectrode platinum tip was measured, and if necessary, was recut to a length of 0.h centimeters (17). Prior to the oxygen diffusion current measurements the standard medium was stirred slowly for 1/2 hour with a paddle stirrer to equilibrate oxygen between the liquid phase and the atmosphere. Ten microelectrodes were then suspended in the medium and the 5 minute diffusion currents determined. All determinations were made at room temperature. Temperature variations between calibration dates were on the order of i 50F, which caused some variability in results over the summer (See Section Iris), but not on any given date. In excess of 500 such determinations were made in 1957. Of these, h63 were considered to be "normal" oxygen diffusion currents. The remainder were abnormally large currents probably due to exposed copper (Figure 1). Such microelectrodes were discarded. The histogram in Figure 2 illustrates the frequency distribution of the h63 "normal" oxygen diffusion currents, representing 266 microelectrodes, some of which were calibrated h times during the summer. More than 80 percent of the measurements fall within the range of the mean i standard deviation. This indicates that the distribution is not normal, but is peaked which is a desirable characteristic of any such calibration procedure. The fact that all the currents were not identical indicates some variability among the microelectrodes. Frequency 68 6h to 36 3 2 28 2h 16 12 12 Mean r 3.5 Std. deviation - 0.30 Std. deviation of the 68 mean t iC.CCLL . in ’ a h? hh P 38 P 21 ' 18 T ' L0 10 ‘1 Y 7 ' 6 S S 1 1 2.; 207 209 301 303 3.5 3v? 3U9 1401 11.3 Oxygen Diffusion Current in Microamperes FIGURE 2. Histogram of L63 oxygen diffusion current measure- ments made in a 3% Na saturated Wyoming Bentonite +0.1 N NaCl suspension; Temperature : BOiE'F. 13 But such variability is insignificant compared with that obtained in actual field oxygen diffusion current measurements which will be discussed in Section II-A. The data show that the selection of "normal" platinum microelectrodes by the calibration procedure described results in a set of microelectrodes with uniform reaction characteristics. h. Wetting of the platinum microelectrodes: Failure of soil mixture to form a complete moisture film on the platinum wire essentially reduces the area of platinum available for reaction with oxygen. Absence of water films stops the operation of the apparatus entirely. Consequently, such a reduction in water film coverage of the platinum reduces the observed oxygen diffusion current and results in an inaccurate measure of the rate of oxygen diffusion. Table II and Figure 3 illustrate the behavior of oxygen diffusion currents in natural soil cores as a flmction of moisture tension. The cores were saturated with tap water and placed on a 10 centimeter tension table for 148 hours. Platinum microelectrodes were inserted into the cores long enough to make the necessary measurements and were then removed. The cores were next placed on the succeeding tension table for another J48 hours. The platinum microelectrodes were again inserted into the cores, using the same holes where possible, and the measurements repeated. This process was repeated until measurements had been com— pleted through 2052 centimeters of water tension. The natural structure of the cores was undoubtedly altered by the introduction of so many holes into the cores; however, since the data were not to be used for m. . rl‘Ili! illllllI-[LE .mcogmndfiopoo oogvooaoonofia m Mo ommnosfix .0; define 05 flats to mgmumé dos m.oa-m.m «mew 0.2 Salem {ma geld? mg. Hind: we 0.9%; 89 gm o.NH-m.m name 042-?“ Nam fled: NA mime; mam mifi unwanméa m4.” moanmd. m.m o.\...w.m 9m o.mnm.m mam Nd envied Nam dmnmd EH fimlmd . m.m finned 8 QN 9T9.” Maw 08qu m4 H.m.a.o m.m "MAIN; 3 0A gain; 5 units HA «.mamé m4 «Auto on as” dmuoé no 31.6 w.o fade 9a. saline 8 min o.~-~.o o4 aééé to H436 FA m.~-m.o 3 5oz owcdm ado: omcdm ado: omedm coo: omcdm Qmm .so olfiiqflma 35M. 5.3 as human! see: oomofihmmf hobodob nowaopco c.3536: monogonofis “359.26 3332 m . mmohpm ohnpmfioz mo 5308.3 w m4 nohou flow Hangmz 5 Esme-oneness antenna cameraman samba HHE Moisture Tension in Cm of water 15 L00 60 L0 30 20 Grayling sand o—o——O Marinesco sandy loam .-n_.-l.. qr Conover silt loam o———0———o LJ Cntonagon silty clay $---9---$ O I a A l A A A A A A 0 2 11 6 8 10 12 111 16 18 Diffusion Current in Microampcres FIGURE 3. Oxygen diffusion current measurements in natural soil cores as a function of moisture stress. lf—Al l6 conquarison of soil structure, these alterations were not an important factor. The two coarser textured soils exhibit a maximum diffusion current. These maxima indicate that the soils were incapable of supporting a complete moisture film on the platinum wire above the tension at the maxima. The measurement is meaningless at tensions greater than that at the maxima. This does not seriously limit the usefulness of the platinum microelectrode method in that at such high moisture tensions in these soils, aeration is generally not limiting to plant growth. The two finer textured soils showed no such maxima. Evidently they are capable of supporting the required moisture film to at least 2 atmospheres tension. It may also be noticed in Table II that, in general, as the moisture tension increases the range in diffusion currents increases. This trend seems to be reversed between 1 and 2 atmospheres tension. Such behavior is indicative of the differential drainage of pores in the soil at the lower tensions, increasing the heterogeneity of the system. .At higher tensions most of the pores are drained and the system becomes more homo- geneous. It may be concluded that the often large range in diffusion currents is not due to erratic behavior of the microelectrodes, but is a measure of the actual heterogeneity of the soil system with respect to the platinum microelectrode. 5. Temperature: Changes in temperature change three of the factors in equation (A) in an aqueous solution, namely the diffusion coefficient of oxygen in .. . . . o a . O .0. . 1 1 1 ‘El‘l'. . . ...1. . .711 . t «w. s 4 1. . . .. . . 2.. .» :- . #11,.llulltll;ylvhlll|lll1li(1_ 1 1.»!al». 4 4.. 11 . Q... ; , r5. .flua . , . ..... , , .l _ .1., -. a s . 1‘1. . . 1 .v . .. 1 stern- on. :0 . 1? time solid-liquid phase (De), the activity of oxygen in the liquid.phase at a gas-liquid phase boundary (GP), and the oxygen diffusion path length (reéR). The diffusion coefficient (De) increases h.3 percent per Centigrade degree increase in temperature (9), while the activity of oxygen and the oxygen diffusion path length (re-R) decrease 1.8 percent and 1.3 percent per Centigrade degree increase in temperature, respectively (5,9). The remaining factors are independent of temperature. Incorporating these changes into equation (A) yields a 2.8 percent increase in the diffusion current per Centigrade degree increase in temperature. In an attempt to determine hOW'Well actual current measure- ment with the platinum microelectrode agrees with data from the literature, oxygen diffusion currents were measured in l and 5 percent potassium chloride solutions at various temperatures. Care was taken to avoid convection by immerSing the test solution entirely in a water bath. Hewever, due to density gradients set up by the reaction at the surface of the platinum, it is certain that the resulting diffusion currents were in some part due to convective flow of oxygen. The convective flow will be directly proportional to the amount of reaction occurring at the platinum surface and inversely proportional to the viscosity of the medium. Relative corrections for convection flow based on a value for 20°C may then be made for the data: . . i - i 1c a 1u - k ( uizoogooc) ( EEgEQ—9 (E) where: iC 1u k i2000 n2000 nTOC . 18 current at temperature T corrected for convection current at temperature T not corrected for convection proportionality constant current at 200C viscosity of water at 2000 viscosity of water at TOC Table III presents the experimental data concerning the effect of temperature on the oxygen diffusion measurement. .Assuming a proportionality constant (k) of unity, the average temperature dependence of 2.8+ percent agrees with the value of 2.8 percent obtained from data from the literature and equation (A). Therefore, in a continuous aqueous phase, the temperature dependence of the oxygen diffusion current measured with the platinum microelectrode is a definite, known quantity. TABLE III Temperature Dependence of the Oxygen Diffusion Current ‘— Uncorrected Viscosity ‘Corrected Temperature Tgmp. 5 Minute of Water, 5 Minute Dependence C Medium Current, [vamp centipoise Current, f7 amp Percent/QC 20 1% KCl u.1 1.005 u.1 3.1 25 1% KCl 5.05 0.891; 11.8 2.7 30 1% KCl 6.15 0.801 5.5 20 5% KCl 3.9 1.005 3.9 30 5% KCl 5.6 0.801 5.1 2.7 Average - 2.8+%/OC 19 In porous media the diffusion coefficient of oxygen and the oxygen activity in the liquid phase determine to a great extent the rate of owgen diffusion toward the platinum microelectrode. The temperature coefficient of these factors would be similar to that for the continuous liquid phase even though the diffusion coefficient is probably not numerically the same in the two systems. Due to the solid structure of the porous media (soil), convective movement of oxygen should be neglig- ible compared to that in a continuous liquid phase. At the present time it is not possible to predict the behavior of the diffusion path length (re-R) with changes in temperature, but since (re-R) appears as a naturfl logarithm in equation (1.)), the effect of its temperature coefficient on diffusion current would be small (0.3%/°C in the continuous liquid phase). Therefore, for most experimental work in porous media the same temperature dependence for the omgen diffusion current will apply in the porous system as in the continuous liquid system. B. Chemical Factors The individual chemical factors considered in this paper are: l. Polarization of the reference cell: The term "polarization" may be defined as "any change produced at an electrode by some means which causes the single oxidation potential to differ from the normal or reversible value in that particular solu- tion" (21). Concentration polarization results during a reaction when the concentration of ions undergoing electron transfer in the immediate 20 zone around the electrode is being depleted at a faster rate than replacements can be supplied from the bulk of the solution. .chhibald (1), using a conventional size saturated calomel reference cell in conjunction with 10 platinum microelectrodes, found that the resistance of the reference cell increased with time during the 5 minute diffusion.period. This indicates a depletion of chloride ions in the zone adjacent to the mercury; Such a concentration polarization results in inaccurate and unreliable data because the internal resistance of the system and the reference cell oxidation potential are not constant. The silver-silver chloride reference cell used at present (17) has 320 square inches of silver surface. This large area reduces the flux density of the reaction in the cell well below that necessary to cause concentration polarization of the cell. 2. Poisoning of the platinum microelectrodes: The term ”poisoning" denotes a chemical change at the platinum surface such as the formation of an oxide coating or the adsorption of some foreign material. It has previously been pointed out that the diffusion.current characteristics of the platinum microelectrodes used during an entire summer remained constant. Each of these microelectrodes was inserted into the soil only long enough to permit one diffusion current measurement and was then removed. The constant stabbing into the soil evidently prevented any such "poisoning" by continually "scrubbing" the platinum surface. Archibald (l) allowed microelectrodes to remain.in.place in the soil over an 8 week period. .At the end of this period the precision of 21 the diiihsion measurements was not as good as when the microelectrodes were first inserted into the soil. Bertrand and.Kohnke (2) allowed the microelectrodes to remain in the soil for a 5 week period. .Although “their corn.plant growth correlated well with oxygen diffusion rates measured with a modified Raney tube, there was no such relationship between.platinum microelectrode data and corn growth. It is probable that this failure of the platinum microelectrode was due to "poisoning." There is no precise explanation for this behavior. However, on the basis of such observations, it is recommended that the platinum micro- electrodes not be allowed to remain in the soil for an extended period of time. The ideal procedure is to insert the microelectrodes into the soil prior to each diffusion measurement. 3. pH: The concentration of hydrogen ions in the diffusion medium may have a marked effect on a polarogram of oxygen (8) and consequently on the oxygen diffusion current measured at a given applied potential. .An experiment was designed to show the effects of pH on.oxygen diffusion currents in two soil susPensions. The first was the standard 3 percent sodium saturated wyoming bentonite plus 0.1 N sodium chloride suspension,“ previously described, which has a pH of 8.0. This suspension was titrated with hydrochloric acid to give the bentonite suspension pH values of h.2, .5.5 and 8.0. The second was a soil sample from a Cecil clay surface material made into a 50 percent suspension and titrated with sodium hydroxide to pH values of 3.9, 5.8, 6.7, 7.8 and 11.0 for individual polarograms. The polarograms consisted of the oxygen diffusion 22 (nxnrent after 1 minute at the designated applied potential between the reference cell and the platinum microelectrode. The suSpension was stirred with the microelectrode between measurements to obtain equi- 'valent conditions for each determination. The data are presented in Table IV and Figure h. pH has little or no effect on the limiting current at a given potential in the bentonite suspension between a pH of h.2 and 8.0 or in the Cecil clay suspension between.pH 5.8 and 7.8. Below a pH of h.2 the observed current is dependent upon pH as well as oxygen diffusion at a given applied potential. This is no real limitation on the method in that very few natural mineral soils have a pH as low as h.2. Above a pH of 8.0 the observed current is again pH dependent. This may prove to be a limitation in the saline and alkali soils of the western United States. Between applied potentials of -0.50 to -0.70 volts there is little change in oxygen diffusion current within the pH independent range of oxygen diffusion currents of pH h.2 to pH 8.0. .An applied potential of -0.65 volts is recommended within this pH range. Deviation in potential of i0.05 volt from -O.65 volt will not appreciably affect the oxygen diffusion current measurement. This is an amendment to procedures followed in previous oxygen diffusion studies with the platinum micro- electrode and to most of the data reported in this paper in which an applied potential of —0.80 volt was utilized. Such data obtained using -0.80 volt potential are not necessarily invalid, but are much more susceptible to small changes in the applied potential and to differences TABLEIV Oxygen Diffusion Polarograms as a Function of pH 23 Applied Potential (—volts) 0.1 0.2 0.3 . . 0.1. 2.6 2.3 2.1. 2.1 1.5 3.0 3.1. 3.0 0.115 3.11 - - - - — -, - 0.5 11.9 2.3 2.5 2.5 - 3.5 3.5 3.5 0.55 6.2 - 2.5 - - - - - 0.6 8.0 2.11 2.5 2.5 1.8 3.6 3.7 3.6 0.65 8.7 2.7 - - - - 3.8 - 0.7 9.2 3.0 2.7 2.5 2.2 3.75 3.9 3.8 0.75 9.11 11.0 3.1. 2.9 - 3.9 11.0 11.0 0.8 12.0 5.5 11.8 11.0 2.5 11.11 11.5 11.1 0.85 - - - - 3.1 5.2 5.h h.8 0.9 - 10.7 8.5 7.3 7.6 6.7 7.1 6.5 1.0 - — 16.71115 19.5 13.0 13.3 15.5 Spot Checks ,0.2 - - - - - 1.5 1.5 - 0.11 - - - - - 3.0 3.2 3.0 0.5 -1- 2.3 2.5 - - - - _ 0.6 - - 2.5 - - 3.7 3.8 3.6 0.7 - 3.h 3.0 2.8 - — - — 0.8 - ~ 11.6 11.11 - 14.3 11.5 11.11 Current in Microamperes ') 8.0. P I0 ?.0 . 6.0 . 500 D 1‘00 P 300 b 3.0 flflflflflflflfl 0"- ‘ 0,,,”"' pH- in? e———e———e ,/ Wyoming Bentonite pH = 5.5 .———-—O—O 7 0 pH i: 8.0 O__..o———O L. pH ~ 3.9 o-—-o-—-o Cecil Clay p” " 5'3 9-"9-"9 PH (,8 O---O—--O pH =ll.O O-—-0-~-0 0.0 _ 3 A L A A A -0.2 —0.3 -0.1. -0.5 —o.6 ~0.’? -0.8 -0.9 F ICURE 1. . Applied Potential in Volts Oxygen polarograms as a function of pH in 3% Myoming Bentonite suspension and in 5C% Cecil Clay suspension. ’- Q"..-.lfW‘_ . “ -I 25 in soil resistance, to be discussed in the next section, than measure- ments made at —0.65 volt. Table IV also reports the results of spot checks of the individual polarograms. In the wyoming bentonite suspension the data show the curves to be highly reproducible. The Cecil clay suSpension, consisting of sand, silt, clay and small aggregates is not as homogeneous as the bentonite suspension.and consequently the polarograms are not as reproduc- ible as in the bentonite suspension. .Also, since steady state is not reached in 1 minute, if the medium.adjacent to the platinum is not representative of the bulk of the suspension with reSpect to oxygen activity, the resulting currents will not be representative of the overall conditions. Such a situation may well occur following a 1 minute determination if the medium adjacent to the platinum is not completely replaced with fresh suspension by stirring prior to the next 1 minute determination. h. Electrical resistance of the soil solution: In the electrolytic reduction of an uncharged substance it is necessary to have an inert electrolyte in the system to provide electrical continuity. .Any factor which results in an excessive resistance between the reference cell and the platinum microelectrode may become a limiting factor in the oxygen diffusion current measurement. Examples of such causes in soils are insufficient electrolyte, physical dimensions of ‘the diffusion system, and discontinuities in the moisture films. Table V summarizes the data of an experiment designed to show the effect of resistance through the reference cell to the platinum 26 microelectrode on the observed oxygen diffusion current. Standard “homing bentonite suspension was placed in each of two beakers. .A glass siphon filled with the suspension served as electrical contact between the beakers. The reference cell was placed in one beaker and the micro- electrode in the other. Oxygen diffusion currents were recorded continu- ously on a Sargent recorder (0-25 microamperes full scale) for 5 minutes at each resistance at each of two applied potentials, -O.80 and -0.65 volt. Resistance was changed by varying the diameter or length of the siphon. TABLE V Effect of Electrical Resistance Through the Reference Cell to the Platinum.Microelectrode on the Oxygen.Diffusion Current Applied Current,_microamperes Potential Resistance Time: (avolts) (ohms) (sec) 2 10 30 60 300 0.80 2,700 - 16.2 8.6 7.05 u.8 6,800 - 10.95 6.85 5.7 b.25 15,700 19.7 10.5 6.15 h.9 3.1 29,800 15.1 9.95 5.9 h.5 3.0 112,500 11.05 9.2 6.7 11.85 2.9 0.65 2,700 '2u.0 10.0 5.9 h.6 3.0 6,h00 20.3 9.6 5.7 h.5 3.0 15,700 16.0 9.h5 5.75 h.5 2.95 29,800 11.8 8.3 5.6 h.h 3.0 h2,SOO 9.h .7.3 .5.h 8.35 3.1 27 In all cases as the resistance increases, the current at 2 seconds decreases. This is due to the limiting current-resistance product which is a function of the applied potential between and the chemical e.m.f. of the two electrodes. However, at the end of 60 seconds the currents are essentially independent of resistance when —0.65 volt is applied, which indicates that up to h2,500 ohms the oxygen diffusion current at the end of 1 to 5 minutes is independent of resistance. 'When ~0.80 volt is applied such is not the case, since, below 15,700 ohms the 5 minute current is resistance dependent. At and above 15,700 ohms the 5 minute current is resistance independent and is equal to the 5 minute current when -0.65 volt is applied. These data are another indication that a potential of -0.65 volt should be used instead of the -0.80 volt proposed previously. The explanation of the results of this experiment is not known at present, and further study is recommended. In the soil, the effects of resistance greater than h2,500 ohms may influence the oxygen diffusion current. However, such resistances are generally not found at moisture contents that will support a complete moisture film on the surface of the platinum wire. For example, in 1957 a great many resistance measurements in field soils were obtained. Only in those surface layers that were approaching the wilting range did the resistance approach h2,500 ohms (i.e. 12 percent moisture in a clay loam and 8 percent moisture in a sandy loam). .At these moisture contents, soil aeration is not limiting for plant growth. It appears that, in ggeneral, soil resistance will not affect the oxygen diffusion current measurement, nor will it limit the application of the method for 28 characterizing soil aeration conditions where such conditions may affect plant growth. 5. Interference from other reducible substances: Substances which are reduced at the same potential or at a more positive potential than that needed to reduce oxygen would interfere with the oxygen diffusion determination by "adding" to the observed current. Lemon and Erickson (12) quote a reference to the effect that in a wide range of soils, no substance other than oxygen was reduced in the soil solution in a polarographic determination for oxygen. Kolthoff and Lingane (8) list the halfawave potentials for a great many cations and organic compounds or families of compounds. The cations that might interfere with the oxygen diffusion current measurement are Co+3, Cu+2, cu+1, Fe+3, MnIZ, and Mo+4 (Table I). Of these, only iron and manganese might be present in soils in soluble form in quantities large enough to interfere with the oxygen measure- ment. Ten parts per million of soluble iron (+3) or manganese (+2) in the soil solution is less than one-tenth the equivalent aqueous concen- tration of oxygen with respect to electron transfer and diffusion coefficient in poorly aerated soils. Since such a high concentration of iron or manganese is rarely found in mineral soils, these cations will not interfere with the oxygen diffusion current measurement. In general, the halfawave potentials of organic compounds are more liegative than —1.0 volts (with respect to the saturated calomel electrode). Illese substances would not be reduced at the potential used to reduce <33Cygen. Some unsaturated acids, nitro compounds, azo compounds, and 29 organic peroxides would interfere if present in sufficient quantity in the soil solution. It may be concluded that there are few instances in mineral soils where substances other than oxygen would be reduced in sufficient quanti- ties to interfere with the oxygen diffusion current measurement. 30 II. RELATIONSHIP BETWEEN OXYGEN DIFFUSION MEASUREMENTS WITH THE PLATINUM MICROELECTRODE.AND PLANT GROWTH Prior to the analysis of oxygen diffusion rates in the field and greenhouse, the sampling variations that will be encountered should be discussed. A, Field SamplingLVariations Measurement of soil aeration with the platinum microelectrode is subject to two major types of field sampling variations. These are: (a) the large random sampling variations caused by the small sampling volume of the microelectrode, and (b) the selection of the individual sampling area at a given site. Generally the soil is a highly micro—heterogeneous system, even though the same soil system may be considered macro-homogeneous. The intermixture of large and small pores may be easily verified by visual observation. This pore size distribution causes in large part the sampl- ing variation illustrated in Table VI. .Another factor is the unequal distribution of moisture. The soil moisture content at this site was determined gravimetrically on 100 gram samples of soil. Even such macro sampling of the soil resulted in a deviation of moisture content of i2.3 percent by weight (Table VI). This variation in the distribution of moisture accentuates the heterogeneity of the soil system. However, if a large enough number of the micro determinations are made, the vari- éations should average out to yield a measure of the macro or homogeneous sacration conditions of the sampling site. TABLE VI Distribution of 120 Diffusion Current Readings in a Conover Sandy Loam (Moisture Content = 111% i2.3% by weight) 31 Current, microamperes Frequency 1.9-2 .0 3 2.7 1 3.5-8.8 9 h.5-5.h 9 5.5-6.h 21 6.5-7.11 21 7.5-8.h 16 8.5-9.3 16 9.5-10.1 1h 10.5-11.1 L 11.5-12.2 5 12.5 1 Total 120 Mean.= 7.h Standartheviation = 2.2 Standard Error 8 i0.2 Six groups of 120 platinum microelectrode diffusion current readings each were obtained in the early summer of 1957 at a depth of h inches in a Conover sandy loam soil. The macro physical conditions differed ampng the 6 sampling sites, but were fairly uniform within each sampling site. Table VII and Figure 5 present a summary of these data. (listribution of the platinum microelectrode readings for the composite 32 .m masmph up popaoaoa mp pew mngEdm o Haw pom monogamoaofle O.H op pmp:doa mosawb map :0 women mp mademm oppmogEou+ mmermoa Mo gonads Hmpop Mo pdmoaom m u mean» Hwoppmpoogew mammemoaofis O.H pmoammd op potency mpemAHSO mo seen I m ado: mamaemoaofis m.o pneumon op empedop mpdoHMSO mo Gems I N new: mamaEdopoHE H.o op pump mpemaaSO mo new: n H see: H em 6 e e e e e 6 adage? La 37 RV a S m a. m a S “.893 mmeaemem «6 $352 8.8 8.0.... made 18.9.. 8.8 8.9 5.8 mice A we Bram 285$ m.m m.m am 67a _ N.N N.N a4 6; A13 8388a 2883 .i 1.. 6;: a? mu” 0d” amp map 888% e...“ Lie 6.: 1a; .3“ -m.m -mia emcee 5:08.: .i 1. 0m mi” 1: mg m5 we 33 m 88: 1. .1 N6 mad mm; a; m6 we Amwv m e8: I .1 me mad 1: N.» me we flew H 58: 39:8 838m w I P , a m N a oppmogsou Ho .mp4. megapz mamewm Emoq_hpemm pm>oeoo up pmepwppo «scum mmdppmmm omH mo mmaaemm 0 Song mpnohado dopm:MMHn mo hamEEdm Hdoppmppdpm HH> mqde av -x" Frequency 165 150 135 120 105 90 145 30 H \'I 33 Standard Deviation = 2.3 Microanpercs 158 -_7 :w_6_ —2 i. +2 Relative Currents in Microamperes FIGURE 5, Composite histogram of 720 diffusion current readings in Conover sandy loam. n.v. I. 3L1 sample approximates a normal distribution about the mean, since approxi- mately 71 percent of the readings lie within the fiducial limits set by i:G_'from the mean and 95 percent of the readings lie within the fiducial limits set by i 2U"from the mean. Therefore, normal statistics may be applied to data obtained with the platinum microelectrode. It is important to know how many diffusion current readings are necessary to obtain a desired accuracy, or, conversely, what accuracy is to be expected from a given number of readings. If the standard deviation of a sample from a population is known, the average standard deviation of the mean of that sample or of a sample of any other size may be determined. 6 i- “ “F'— In (a) where 621 = standard deviation of mean (standard error) 6— 8 standard deviation of the sample population _n = number of determinations Therefore, to obtain a standard error of $0.2 microampere, it would be necessary to take 120 readings per location in the Conover sandy loam (from Table VI). On the other hand, if only hO readings were obtained, which takes about one-half hour: (if “/1232 I: 30.35 microampére (1) The standard error under these conditions is i0.35 microampere. This is a more practical number of readings per location, and results in a reasonably accurate sampling of the given site. It is possible that different sites will result in radically different standard deviations of'the current readings, and this is a point that must be taken into 35 consideration when sampling. But the Conover sandy loam represents as heterogeneous a soil as was sampled in these studies and the standard error for this soil is greater than for any other site. Therefore, all platinum microelectrode field data presented in this paper represent the average of DO diffusion current readings per sampling site, except where indicated otherwise. I The magnitudes of the mean and standard error remain about constant when the current values are rounded off to the nearest 0.5 microampere or nearest microampere. This coupled with the magnitude of the standard error indicates that possibly the diffusion currents need not be determined to the nearest 0.1 microampere. If the currents were determined to the nearest 0.25 microampere, the size and cost of the microammeter could be reduced, and the Speed with which the readings are obtained could be increased. Therefore, it is suggested that future apparatus for measuring diffusion currents with the platinum microelectrode contain a fairly compact, rugged meter, and that the currents be read to the nearest 0.25 microampere. The selection of the individual sampling areas at a given site can introduce an error common to all sampling techniques for physical proper- ties of the soil. The oxygen diffusion current measurement is particu- larly sensitive to this error. .At a site with uniform surface, bulk density and moisture content, which is the general case in the green- house, such errors will be negligible. However, in any field that has been traveled by a tractor after’plowing or tilling there will be differences in bulk density of the surface soil between where the tractor 36 wheels have and have not been. Tillage implements do not affect the entire surface of the field in a reproducible manner. Concentration of plant roots tends to deplete the oxygen concentration of the adjacent soil atmosphere. Differences in micro—relief will affect the distribution of moisture over the field. When sampling such a field for oxygen diffusion rates, these variations in field conditions will cause marked differences in the results obtained depending upon the location and distribution of the individual sampling areas. To illustrate this point, data obtained in 1957 in the Ferden farm sawdust rotations are presented in Table VIII. On alternating days, operators A and B took 1.10 diffusion current readings in each plot. Operator A placed three—fourths of his microelectrodes between the corn rows where the tractor wheels had passed, while operator B placed three- fourths of his microelectrodes in the corn rows or close to the rows. These data were obtained for several weeks following 7 inches of rainfall. Had the two operators sampled the plots in a similar manner, a general increase in the diffusion currents with increasing time would be expected. That is, as the plots dried out, the rate of diffusion should increase with the increasing air filled pore space. Such an increase my be observed, but it is partially masked by the large variations in results obtained by the two methods of sampling of the same plot. Therefore, in sampling a given plot for oxygen diffusion rates, the microelectrodes must be distributed uniformly or reproducibly over the sampled area. Oxygen Diffusion Currents Obtained in Ferden Sawdust Rotation.Plots in 1957 TABLE VIII 37 “5 Minute Reading, microamperes “rm-313731 Plot 141-132 M Operator Operator Qperator Date A B A B A B 7/15 5.1 5 .0 5 .0 7/16 5.8 6.0 h.8 7/17 5.2 6.0 7 .15 7/18 7.8 8.0 5.6 7/19 6.11 6.2 8.2 7/29 8.11 8.5 7.6 8/5 8.25 8.9 8.1 B. Evaluation of Critical Ppptions of the Growth.Peridd;pf Tomatoes The problem of the critical growth.period has been studied.with tomatoes in the greenhouse and in the field. 1. Greenhouse experiment: In the winter of 1957 an experiment was initiated in the greenhouse to study the effect of a short period of oxygen stress or "shock" on actively growing tomato plants. The tomatoes were transplanted on February 15, 1957 into sandy soil material in cylindrical containers 8 inches in diameter and 2h inches high. plant. allowed to drain from the bottom of the container. Each container held one tomato The plants were watered with Shive solution and the excess was For the oxygen "shock" treatment the water level was brought to within 10 centimeters 38 of the surface of all containers, except the control, for a period of 7 or 22 hours once during the growth period. The oxygen diffusion rates measured with the platinum microelectrode in the saturated soil material [were on the order of 28 i 6 x 10..8 gm/cm2 min in all cases, which was below the limiting value of 30—hO x 10"8 gm/cm2 min for tomato vegetative growth obtained.by Lemon (10). Following the predetermined saturation period the soil material was allowed to drain freely. .A summary of the data obtained is presented in Table IX. There were no consistent differences in plant growth to 7 or 22 hour shock periods as indicated by the results of student t tests. Evidently 7 hours was sufficient to produce any affects that limited soil aeration may effect in actively growing tomato plants. Vegetative growth was hindered only in the early portion of the growing period, and then only when the day was clear and the plants were presumably reSpiring and tranSpiring at a rapid rate. Oxygen shock of the plants at later dates had little effect on vegetative growth regardless of meteorological conditions. The production of tomato fruit was much less uniform than that of vegetative growth, due possibly to the early harvest. Oxygen shock early in the growing period reduced fruit yield mildly, regardless of meteorological conditions. Reduced fruit yield was observed toward the latter portion of the growing period when the temperature of the greenhouse was quite high. 'Within the limits discussed in the introduction, these data lead to the following conclusions concerning the effect of insufficient soil oxygen diffusion rates on the growth of tomatoes: 39 .pmdpmppo hHHmSpom mama.dmep pdospmoap Mooem ado: mm opp mo mpaowh pew pdoEpmoap xooem Ado: 5 map mo mpaoflh domspon moodmpoMMHp popmmaw mdfldpmpno mo hppfipppmmom . pdogma om m mp maoep peep 30cm span pHmHh dop pew ppsam map mo mpmmp p pdmpdpm*** .mpamad u now. mommaopw mam mosawb pmxooem poz .mpdmam opmsop m ho m pom mommaopw mam modamb omega ** .wmma «:H_hmz do popmobnme mpcmam .wmma «ma humsnpmm do popdmammdmap moopdfioew I I1 ix. .mEM1w.p3_mup «mace kl?" “I Irll .mmm “pagan monomwom pdmam amma .meeeeeeeeo was up mmopmsoe apps.pemapnodxm xoonm dmmmxo mo hMMEEBm spawn fleece NH mqde m.pa e.mm . mm eeaeoem poz e.ma 6.6m a.am e.me mas Hm a\m N.aa m.pa N.Hm m.ma map ea om\a «.3 a. 3 ma mam an 8 a} H.6m e.ma a.am e.mm mam a6 mm\a m.Ha e.mm 0.46 o.pm mam mm m\: e.ma m.om e.mm m.mm mom a: wm\m m.ma o.aa 6.pa p.am mom am Hm\m m.6m p.am 6.3m o.pa em: ma mm\~ as 92 ea 33 em a we xooem xooam Mooem xoosm nae m5p\amo 5w whee «wdppdmadmdmae pmaoopm .em Nm .em a .em mm .em a coppeaeem emaom . *eepem mafia memo 110 (1).A period as short as 7 hours of limiting oxygen is sufficient to produce marked effects on tomato growth. (2) Such effects are obtained, in general, in the early portion of the growing season. (3) Such effects are more liable to occur when the plant is most active, such as on a sunny day. (h) These effects result in a reduction in.vegetative growth and tomato fruit yield. These effects are not all inclusive to the plant kingdom and should be applied only to tomatoes. In contrast to the tomato, Cline (3) found the most susceptible portion of pea growth to oxygen stress to be just before flowering and continuing through pea formation. Similar experi- ments on other common crop plants should be performed to determine the effects of sub-optimum soil aeration conditions on plant growth with respect to the duration and magnitude of the deficiency of oxygen, and the most susceptible portion of the plant growth period. 2 . Field experiment: In early June of 1957 a series of experimental plots were estab- lished on a Conover sandy loam soil which.was designed to produce differences in the aeration of the soil and its effects on the growth of tomato plants. .A 2 foot high alfalfa crop was plowed down 3 weeks before transplanting the tomatoes. The soil was treated as follows: 1. One-third of the area was not further tilled. 2. One-third of the area was tilled with a roto tiller. 3. One-third of the area was disked 3 times and rolled 3 times with large water-filled rollers to produce as much compaction as possible. During the growing season water was applied to these plots with over- head type F Sprinklers as follows: 1. No water added in excess of natural rainfall to one-third of the area. 2. One inch of water added per application to one—third of the area. 3. Two inches of water added per application to one-third of the area. The combination of low infiltration rates and/or high rates of water application on most of the plots receiving irrigation made it necessary to pond.water on these plots with small earth dams. .All possible combinations of tillage and moisture were fabricated and replicated h times. This resulted in 36 plots, each 15 feet by 18 feet containing 15 tomato plants Spaced 3 feet by 6 feet. The h plants in the 6 feet by 10 feet sampling area in the center of each plot were selected for study. .After the tillage operation was completed, the surface of the sampling areas was undisturbed except for transplanting of the tomato plants and the insertion of the platinum microelectrode and modified Raney tubes (1h,18) for the determination of oxygen diffusion rates. IMorton's hybrid tomato plants were tranSplanted on June 13 from.wooden planks set on concrete blocks so as not to step on the sampling areas and disturb the structure. .All oxygen diffusion measurements were made ' I "aw-AIM. _ h2 from these planks. 'Weeding was done by either scraping the surface or by cutting the weeds with shears. Oxygen diffusion measurements were taken at a depth of h inches 2h hours after each irrigation on all plots. In order to accomplish the sampling of all 36 plots in 1 day only 20 microelectrode and 2 Raney tube measurements were obtained.per sampling date, thus sacrificing some accuracy for speed. Irrigation was continued into September. Tomato vegetative growth rate was determined by measuring the length of the longest stem on each plant. Tomato fruit was harvested at intervals starting on.August 5 and continuing through.September 27, 1 day after the killing frost. This same day the above ground vegetative growth was removed and later dried and weighed. Tomato roots from selected plots were dug at this time. From October 3 through October 5 core samples were taken from a depth of 2—5 inches, 10 cores per plot sampling area. Due to soil variation and variation in tillage treatments, the groups of h plots which were set up as replicates differed considerably in aeration characteristics, moisture content and bulk density. Therefore, each plot will be treated separately in the analysis of the data. Measurements of vegetative growth rates were discontinued after 1 month. By this time some of the plants were no longer standing upright and length measurement became difficult. .At the end of this period there were no differences in length of plants between plots. However, the total amount of foliage was visibly different among some of the h3 plots in which the tomatoes had equal length of stem. It may be con— cluded that such a method for measuring plant growth rate is not well adapted for total vegetative growth. Large differences in total vegetative growth, root growth and tomato yield are readily evident (Table X). These differences in yield may be due to a number of causes, namely fertility, disease, high bulk density, high or low moisture contents, and lack of sufficient gaseous . , «1 artv'fi I p. 5 exchange in the soil. “"1. The plots were not fertilized in 1956 or 1957 except for a small 331'— 1 $5"?! Sui“: '- ' amount of starter solution added around the roots of the tomatoes when they were transplanted. The variations in natural fertility could have caused small differences in yield, but not such large differences as actually occurred. The plants appeared to be disease free, at least in the above ground portion. High bulk density is not the cause in itself in that for plots with high surface bulk densities and low early season moisture content (plots 833 and 111) the tomato yield was high, while in other plots with high surface bulk densities and high early season moisture content (plots 131 and 121), yield was quite low. High moisture contents in themselves were not the cause for low yields. In plots with high early season moisture content and low surface bulk densities (plots 632 and h32) the yield was good, while in the plots with high surface bulk densities and high early season moisture content the yield was poor. One of the plots (512) had such a low average moisture content that this is believed to be the limiting factor for yield on this particular plot. 1.1-11-4.pz ND 10mg . CHE mEo\mIOH N 511.-.de >3 _¢Q\Q; .dfl: mao\m10fi on So m8d\§.-1§1d§1dAd1fi1-¢2 ...... 28362 85 Bee eon-due 8336: med 8.8 copes-EB append. pd poem e5 18$ page ”BE pdoopmm modem mooapooaooaoflz .pm pdooaom hoemm ooompooaoopofiz .pm Masm paoww omeoe RENE 8. Qa- «meeme SEE op ab 489.8 deHaEmm m pop wpmo mo owmpo>4 mapademm Q How mpmo Mo owmpobd. wmma .Epmm hppmao>ficbopmpm dmwfleoflznpdeHaodxm doppmwod.opmsoa opp mom moHopw cpmsoe odd Hpom map mo moQQCH o>pm conundm map pop open Hmopmhnm mo hamEEdm HUMAmay TABLE X Summary of Physical Data for the Surface Five Inches of the Soil and Tomato Yields for the Tomato Aeration Experiment,Michigan State University Farm, 1957 Average of data for 8 sampling Average of data for 3 sampling dates, 7/9 to 9/6/57 dates, 7/9 10 7/26/57 Tomato Yield Bulk Pt. Microelectrode Haney Percent Pt. Microelectrode Haney Percent Plot Fruit Vine Dry Root Dry' Density diffusion rate Tube Moisture diffusion rate Tube Moisture No. lbs. Wt.,gms. Wt.,gms. gm cm3 gm x 10’8/cm2 min D/DO by wt. gm x 10‘8/cm2 min D/DO by wt. 131 20.7 361 1.17 16.5 0.15 13.1 329 0.18 18.2 512 37.9 675 1.18 56 0.61 9.6 53 0.19 12.1 833 61.1 1078 1.12 53.5 0.60 11.8 15 0.56 13.7 1021 75.8 1150 1.38 53.5 0.52 13.1 17 0.105 17.3 111 62.1 761 1.50 36 0.17 13.2 12 '0.125 13.2 522 11.5 507 1.11 15 0.15 15.7 39% 0.12 15.1 813 68.1 1219 1.12 51 0.50 11.9 11 0.51 15.1 1031 69.3 162 1.37 52 0.19 17.6 19 0.115 18.3 121 26.7 233 21.3 1.52 35 0.22 17.6 30w 0.16 18.1 532 51.0 101 55.8 1.16 37 0.30 17.8 35 0.33 17.2 823 60.7 518 55.7 1.11 16 0.13 16.1 16 0.15 11.9 1011 37.2 116 18.5 1.39 11 5 0.31 18.8 10* 0.22 19.8 231 25.6 509 1.31 15.5 0.18 15 8 37* 0.13 18.3 612 81.6 1518 1 23 17 0 61 12 1 51 0.19 15.2 933 55.9 958 1 10 53 0 72 15 2 51 5 0.68 16.9 1121 83.5 1550 1.21 18 0 69 12 1 51.5 0.61 17.1 211 38.6 676 1.39 15.5 0.19 . 16.2 13 0.11 18 1 622 91.8 981 1.19 13.5 0.56 19.5 . 17.5 0.51 19.9 913 56.6 831 1.37 50.5 0.18 15.7 , 19 0.18 16.7 1131 82.8 877 1.28 15 0.18 , 18.8 19 0.11 19.8 221 38.2 612 1 15 36 0 21 19 2 351 0 26 1 o o I\ o 9.9 632 76.6 850 1.28 13 5 0.36 20.0 16.5 0.11 19.1 923 53.1 018 1.31 . 16 ‘ 0.59 17.7 19 0.62 17.6 1111 79.9 810 1.36 12 0.30 18 8 13 0.295 20.7 331 62.6 1110 61.1 1 20 17 5 0 69 1 , . . . . 3.55 13.5 0.6“ 18.0 112 E51 926 62.0 1.38 15.5 - 0.60 10.7 15 0.55 13.9 123i 8%.1 511 71.1 1.30 50 0.75 11.85 18 0.71 13.8 -8 1296 79.0 1.25 19.5 0.66 11.8 19 0.19 15.9 311 35-0 573 l 11 12 5 - O 11 15 - -, 3 37 . 0. 6 16. 3123 53191151 1:930 1.27 13.5 0.52 17 1 13.5 0.56 18.8 1231 8 - 7 1.35 50 0.58 15 25 52 0.61 15.1 7.3 773 1.33 16.5 0.55 16 9 17 0.53 17.8 ii; 11': 3%; 1.31 37 0 16 19 0 35 , 0.51 18.7 723 90-2 108 1.29 11 , 0.11 19.0 11 0.18 19.0 1211 87.5 818 1.22 12.5 0.52 18.0 13 0.56 16.9 . 1.32 15.5 0.12 17.2 16 5 0.395 17.6 ‘Correlation coefficient between tomato fruit yield and o ' ' _ _ . _ Egalwfflmumrat l 0.19f :Easphplgt:l:1tpldiffupion rates equal to or less than the critical rate of6 efigagE/cmg minor 0 1e in e ' b d d b ' ' ' ' coefficient at the 21 peroggiligvelugqialsyOEEISTOtted lines in Figure 8. Correlation 1r 1:- /‘¥‘!!.!§\Y 15 The remaining factor in tomato growth is soil aeration. In order to study the effects of aeration on tomato yield, it is necessary to know which of the data obtained are most applicable to such a study. (In other words, should an average figure for the aeration condition of the plots over the entire season he used, or should the data from only a portion of the growing season be used? Based on the results of the greenhouse experiment on oxygen shock of tomato plants, the answer is that oxygen stress in the early part of the season is more critical than at a later date. .Analysis of the data obtained in the field experiment substantiates these conclusions. Figures1 6 and 7 are scatter diagrams relating oxygen diffusion rates obtained with the platinum microelectrode to tomato fruit and vine yield. Each point on these graphs represents the per plot average of diffusion rates taken 21 hours2 after irrigation on 8 dates ranging from 26 to 85 days after transplanting the tomatoes. Figures1 8 and 9 are similar diagrams for data obtained on only the first 3 dates described above, or from 26 to 13 days after tranSplanting the tomatoes. There is very little, if any, relationship between oxygen diffusion measurements by either method and tomato response if the data for the entire growing season is used. However, the scatter diagrams for the data from the early portion of the growing season indicate a definite relationship between these factors. Below an oxygen diffusion rate of lData from'which the graphs were made are in Table X. 2Determinations on the first date were actually 7 days after heavy'rains. ‘1‘_»1 1 .=~ ) «.Lz‘fl- ‘. Irma *3 ‘ +J C) r ‘4 Cl. $4 CJ :1. (7) '9 L. .3 0 Cl. q "C .4 :11 61 >« + J -H 73 5.4 L... n 4.) 0} §:. 0 H 90 YO 50 10 30 1.16 C P . . 0 o O C o O C 1 o I O C O C h C O O O O p C O . . ' o C C L C O . . C O O A l I . l L 28. 31 10 16 52 58 5 Minute Diffusion Rate (gm x 10‘8/012 min) FIGURE 6. Scatter diagram of the oer plot averages of oxygen diffusion rates taken 21 hours after irrigation on 8 dates ranging from 26 to 85 days after trans- planting the tomatoes. Tomato Vine Dry Weight, Grams per Plot '1600 1500 11.00 1300 l2OO 1100 L030 900 800. 700 600 500 1400 300 200 100 L l l ’ l L 28 FIGURE Y. 3h b0 L6 52 58 5 Minute Diffusion.Rate (gm x 10‘8/Cm2 min) Scatter diagram of the per plot averages of oxygen diffusion rates taken 2h hours after irrigation on 8 dates ranging from 26 to 85 days after trans— planting the tomatoes. l:- .l ‘ ‘ Is ‘6. ‘.‘AM .. Tomato Fruit Yield, Pounds per Plot 90 YO So 30 I I O I .. I O I I I . ’I r I . O O I | I I. " I I I C _ I ’ . I I . I I I I , . I , I '- I I I I . . o/ I ° . // O I/ / P / / / I/ / /’ / ' / /' ./ . / / 0/ °/ 0 /' o ./ / / / ° // ./ P /' /,/ _/ / /. o/ ‘ ./ ./ o A/ 1 A A 28 3h to us 52 58 5 Minute Diffusion Rate (gm x W‘s/cm2 min) FIGURE 8. Scatter diagram of the per plot averages of oxygen diffusion rates taken.2h hours after irrigation on 3 dates ranging from 26 to a; days after trans— planting the tomatoes. , Grams per Fist II'IIei ght Pry omato Vine '71 J 900 800 (00 600 500 too 3620 2’00 100 / I / .. I . / / I / I P / i / I- / . .l / I 4 I b / / . I / , I ,. / o o I / . / ' / ' .I / o | r / . I / ' . I I' / o o / .0 O / / / . //' . // . ‘/ _ /o. / / o /’ / 9’ o p / O O ./ / . // O / /’ I- / . / /’ / ' // ./ / / . /’ ,. / / / / 98 3h . ~ ho D6 - S? 58 5 Minute Diffusion Rate (gm x iO‘e/cm‘a min) FlGURE 9. Scatter diagram of the per plot averages of oxygen diffusion rates taken 2h hours after irrigation on 3 dates ranging from 26 to h} days after trans- planting the tomatoes. So to x 10"8 gm/cm2 min as measured with the platinum microelectrodes, tomato fruit yield and to a lesser extent tomato vine yield and tomato root yield, within the limits of the scanty root data, are reduced by lowering the oxygen diffusion rate in the liquid phase. The correlation between fruit yield and diffusion rate below the "critical" diffusion rate is significant at the 2h percent level. This lack of significance may indicate that there is a range of diffusion rates in which the plants . -v_._,-n_1 “— may be more or less affected by limiting oxygen. Below this range may be a region where yield and diffusion rates are better correlated. The same trend is evident from the modified Raney tube data. Since these tubes measure the rate of oxygen diffusion in the gas phase in the soil body, it is evident that the rate of oxygen diffusion in the gas phase as well as in the liquid phase is important for plant growth. Above this critical diffusion rate tomato yield is not affected by aeration conditions in the soil. Hanks and Thorpe (6) observed the same effect of oxygen diffusion rate measured with the platinum microelectrode on the germination of wheat seedlings. Below an oxygen diffusion rate of 75 to 100 gm/cm2 min the percent emergence of the wheat decreased rapidly; .Above this critical rate the percent germination was not influenced by increasing the oxygen diffusion rate. The observed yield variations above the critical oxygen diffusion rate may be due to the factors affecting plant growth mentioned previously in this section. Subject to the restrictions on the applicability of these oxygen diffusion rate measurements as discussed in the Introduction, 51 these data lend reasonable support to the proposition that soil aeration conditions characterized by oxygen diffusion rates as measured in this experiment may be correlated with plant growth for low rates of oxygen diffusion. C. Evaluation of the Effect of Soil.Aeration on Crop Yield One practical application of the measurement of oxygen diffusion .‘Lt- fl ‘1 ;. rates is to enable the agronomist to evaluate the role of soil aeration hmh‘_‘ '0- on field crop yield. To do this, critical oxygen diffusion rates for each crop studied must be known, and the oxygen diffusion rate measure— ments must be taken during the portion(s) of the growing period most sensitive to oxygen stress (Introduction). At the present time only rough approximations to the critical oxygen diffusion rates for sugar beets are known. ‘Wiersma and Mortland (20) determined the critical value for sugar beet growth to be 20-30 x 10‘8 gm/cm2 min of oxygen. lArchibald (1) obtained critical oxygen diffusion rates of 30—35 x 10“8 gm/cm2 min for sugar beet germination. .During germination and early seedling growth, beet response was greatly affected by soil aeration conditions. Those beets surviving this period seemed to adapt themselves to the adverse environment and produced "normal" beets. From these observations, Archibald concludes that the first few weeks is the most critical portion of the growth period for sugar beet growth with reSpect to oxygen diffusion rates. In 1955 and 1956 oxygen diffusion current measurements were taken during the first month after planting of sugar beets on the Ferden rotation plots, Chesaning, Michigan. 52 In 1957 the oxygen diffusion current measurements were taken over a similar interval of time following a general 7 inch rain in July. The data are summarized in Table XI. .A marked relationShip between oxygen diffusion rates and sugar beet yield is shown by the data. The diffusion rates in 1955 average 8.5 x 10"8 gm/cm2 min greater than those for 1956 and 1957. The corresponding yields for 1955 are over twice those for 1956 and 1957. The approximation of these diffusion rates to the critical value of 20-35 x 10‘8 gm/cm2 min mentioned.previously for the sugar beets indicates that limiting oxygen diffusion rates during this portion of the growth period of sugar beets may well be the major factor in causing the large decrease in beet yield from 1955 to 1956. It should be stated that oxygen diffusion rates during May and June of 1957 were probably equal to or less than those for a comparable period in 1956 due to an overabundance of rainfall during those months. This latter statement is intended to discourage speculation that the period of time approximately 2 months after planting may be as susceptible to oxygen diffusion rates as any other portion of the plant growth period. In other words, conclusions concerning such critical portions of the growth period should not be drawn from the data presented in this section. Nevertheless, these data do indicate a distinct relationship between oxygen diffusion rates measured with the platinum microelectrode and sugar beet yield. .A similar study was carried out on corn on the Ferden rotation plots. The data are summarized in Table XII. These results differ greatly from “ _-.... Sh TABLEZXII Relationship Between Corn Yield and Oxygen Diffusion Rates Measured with the Platinum.Microelectrode on Ferden Rotation Plots OxygeniDiffusion 1 EIOt Rate, 3 Corn Yield Year Rotation Replicate .gnlleO‘B/cmz min bu./acre 1955 l l h3 70 ‘ 2 h3.5 91 1956 l 1 35 lot 2 3h 116 1955 6 1 h3.5 60 2 h3 67 1956 6 1 36 69 2 3h 62 1Corn planted on May h, 1955 and May 28, 1956. 2Rotation l s 2 years alfalfa brome, corn, sugar beets and barley = 5 years Rotation 6 a beans, wheat, corn, sugar beets and barley - 5 years 3Average value for 2 sampling dates, within 29 days of planting in 1955 and within 23 days of planting in 1956. those of the sugar beet studies. Although the oxygen diffusion rate values for 1955 average 8.5 x 10"8 gm/cm2 min greater than those in 1956, there is no corresponding decrease in corn yield from 1955 to 1956. Rather, there is a large yield difference between rotations 1 and 6, regardless of the year or diffusion rates. Bertrand (2) obtained a .w‘ g“? rm w SS critical oxygen diffusion rate equal to or less than 25 x lO'Bgm/cm2 min for corn growth. This indicates that the oxygen requirement of corn is very low. It is quite apparent that corn is not affected.by oxygen diffusion rates under conditions similar to those that appear to produce large changes in sugar beet yields. D. Use of Rate of Changg of Oxygen Diffusion Rates FollowinggRainfall or Irrigation . In the event of rannfall heavy enough to saturate or nearly saturate i the surface of the soil, gaseous transfer between the soil atmosphere paw-3'; 5.0.3. and the atmosphere above the soil is greatly impaired by water clogging the pore spaces. This would be expected to result in a decrease in the oxygen content and diffusion rates throughout the profile if there were any processes occurring that demanded oxygen. The duration of such a condition depends largely upon how rapidly the soil can dissipate this excess water and restore the original gaseous exchange routes. If the magnitudes and rates of change of oxygen diffusion rates were measured following such a rain, it might be possible to estimate the portions of the growing period of a field crop during which.oxygen diffusion rates were below the critical value. Some of the information necessary for this purpose are the critical oxygen diffusion rate for the crop, knowl- edge of the rainfall pattern for the growing period, and assumption that the physical conditions of the soil when the oxygen diffusion measurements were obtained were representative of those conditions during the other rainfall periods. 56 Provided a reliable estimate of the duration and distribution of oxygen stress periods can be obtained for a given growing period, the agronomist may be able to estimate the influence of oxygen diffusion rate on the growth of the crop in question. Then if soil physical conditions are found to be conducive to poor aeration, the agronomist may be able to take steps to relieve this situation. In this manner the ability to accurately measure the oxygen diffusion rates in the soil and relate them to plant growth should be an important contribution to the science of agronomy. In 1957 the measurement of changes in oxygen diffusion rates were determined under 2 sets of circumstances. On July 8, 1957 a h inch rain fell on the Ferden rotation plots, followed by another 3 inches 3 days later. Oxygen diffusion rates in the sugar beets of rotations1 l and 6 were measured.with the platinum microelectrode at intervals following the heavy rains. The data are summarized in Table XIII. In the second case irrigation was used to substitute for rainfall. Portable type EA sprinklers (h) were set up in the desired location on Ferden rotation corn plots and approximately h inches of water was applied over an area 2h x h2 inches. Diffusion measurements were taken following the irrigation, and a summary of the data is presented in Table XIV. The data for oxygen diffusion current change following a heavy rain in sugar beet plots reveal that it took at least 10 days for the diffusion rate to exceed the critical value of 20-35 x 10"8 gm/cm2 min 1For description of the rotations see footnote 2 to Table XI. TABLE XIII Change of Oxygen Diffusion Rates in Ferden Rotation Sugar Beet Plots Following a General 7 Inch Rain 57 Rotation Replicate After Rain After Rain Plot Oxygen Diffusion Rateifl 6’ Days III Days 21 Days x 10‘8/cm2 min '28 Days After Rain After Rain l 1 ZI-I 33 37 36 4 2 2h 28 36 Id. 3 28 32 LIE-5 M. II 26 30 II]- M Average 26 31 I40 _ I10 .5 6 1 22 26 33 .5 39 2 25 .5 28 II3 .5 I42 3 25 28 I41 . I13 .5 II 2II 28 II3 __ 51 Average 24 27 .5 I40 IIII TABLE XIV Change of Oxygen Diffusion Rates in Ferden Rotation Corn Plots Following Application of II Inches of Water with the Type FA Sprinkler Oxygen Diffusion Rate, gm x 10"8/cm2 min Plot Prior to 2h Hours After 72 Hours After Rotation Replicate Sprinkling Sprinkling Sprinkling 1 3 36 25 35 II 38.5 30.5 28 Average 37 28 31 .5 6 1 II? 22 3h 2 27 2o 25 3 26 2h 28 u h2 21 f -- Average I40 .5 22 ~ 29 ' 58 for sugar beets. Thus for 10 consecutive days the sugar beets were in a state of oxygen stress. Even if this portion of the growth.period for sugar beets were not found to be as critical with respect to growth and yield as other periods, such prolonged oxygen stress should certainly have a deleterious effect on the production of sugar beets on those plots in 1957. "r Hf? ' Since very little information concerning the relationship of corn 3 growth and yield to oxygen diffusion rates is available, the data for the Ferden rotation corn plots are presented solely to demonstrate the u—v—u—r‘ gnu“. changes in oxygen diffusion that may occur following artificial irrigation as described previously. The oxygen diffusion rates decreased after the addition of the water and after 72 hours had started to approach the rates prior to the irrigation. These trends are to be expected. The magnitudes of the rates and the speed of the increase in rates are two of the factors that may eventually be used to evaluate the effect of oxygen stress on crop plants. E. Discussion With.respect to the field and greenhouse measurements reported in this paper, all of the necessary data for the most accurate oxygen diffusion rate-plant growth correlations were not obtained. Sufficient meteorological information is available for such an analysis. In 1956 and 1957 soil temperature measurements were made in the field. However, the method used to measure the oxygen activity in the soil solution has since proved unreliable, and this failure prevents a calculation of the ——-—4 5““ A , S9 cnqvgen diffusion rate-plant growth correlation under the conditions Specified in the Introduction. Therefore, the data presented in Section II represent oxygen diffusion measurements taken at all times throughout the day with temperature, oxygen activity and meteorological conditions not considered. This may explain the failure to obtain highly significant diffusion rate-plant growth correlations. Such data may be used only to show the general relationship between oxygen diffusion rate measurements with the platinum microelectrode in the field and greenhouse and plant growth. The method used to obtain a measure of the oxygen activity in the soil solution has frequently been recommended to other investigators as a valid measurement. This error should be explained. The main assumption involved is that the activity of oxygen at the surface of the platinum microelectrode prior to any reduction of oxygen is directly proportional to the activity of oxygen in the body of the soil solution, provided sufficient time is allowed for establishment of an oxygen equilibrium between the two phases. Data obtained from the literature seemed to support this assumption. When a potential of —O.65 volt is initially applied between the platinum microelectrode and the reference electrode, the oxygen reduced should be that adsorbed on the platinum surface. The observed current should be directly proportional to the adsorbed oxygen and thence directly proportional to the activity of oxygen in the body of the soil solution. .An.ordinary DC microammeter with a'DKArsonval movement was used in place of a ballistic galvanometer to measure this initial or A ‘ ‘- ”I m .‘a “instantaneous" current pulse. The current is continuous and decreases rapidly with time during the first few seconds, but the meter movement recorded a maximum current as if it were a single pulse. Various methods were used to determine the relationship between this maximum meter read- ing and the actual magnitude of the current close to zero time. Not until 1957 when a highly accurate recording potentiometer became available was an attempt made to determine the experimental relationship between the maximum ”initial" meter reading and the oxygen activity in the soil solution. For a 10 fold change in oxygen activity, less than a 2 fold change in the maximum meter reading was observed. In addition to this very low sensitivity the maximum meter reading was found to be greatly influenced by the ionic strength of the soil solution and/or the resistance in the diffusion medium between the platinum microelectrode and the reference electrode. Therefore, up to the present time, all data from this method of measurement of the oxygen activity in the soil solution are probably not reliable. It is possible that in the future this measure- ment may be improved, but until that time, this method for oxygen activity measurement is not recommended. Reference has been made throughout this thesis to oxygen diffusion measurements with the platinum microelectrode and their relationship to plant growth as obtained by several other investigators. A more complete description of their experimental conditions and results is presented in Table IV along with results obtained by this investigator. All these results are subject to the restrictions placed on this investigator‘s results in the opening paragraph of this section, namely lack of soil TABLE XV Summary of Critical Oxygen Diffusion Rates as Measured with the Platinum Microelectrode Oxygen Diffusion Rate Portion of Most Critical Portion Growth Sampling Depth Critigal Rate Plant Growth of Plant ' Investigator Crop Conditionsl (inches) gm X lo 8/cm2 min Period Studied Growth Period? Lemon (l0) Tomato Greenhouse, 8 30—h0 First 5 weeks 75 F temp., after trans— ——— Winter planting Van Doren Tomato Greenhouse, h at least 28 i 6 for l2 weeks within 2 weeks Winter after trans— after trans- planting planting Van Doren Tomato Field h cf hO entire growing within 7 weeks Summer season after after transplanting transplanting Archibald (1) Sugar Beets Grgenhouse, 2 and 8 30—35 first 9 weeks seedling 80 F temp., from seed stage Winter Wiersma (20) Sugar Beets Greenhouse, h 20—30 first l2 weeks Spring from seed ——- Van Doren Sugar Beets Field, h at least first h weeks Spring 30—35 from seed —-~ Cline (3) Peas Grgenhouse, 3 CK 7O entire growing from flowering 60 F temp., period from to pea ‘Winter seed setting Hanks (6) 'Wheat Greenhouse l/2 75~lOO seedling stage ——~ Jackson (7) Potatoes Greenhouse 2 30—50 first 5 weeks . ,7 80-100 F temp., from sets Summer ; Archibald (1) Cats Grgenhouse 2 and 8 < 30 first h weeks 1 80 F temp. from seed ——- Spring Bertrand (2) Corn Greenhouse, 8 to l6 < 25 first 5 weeks from seed ——— van Doren Corn Field, h < 35 first h weeks Summer from seed 1Relates the general conditions in which the test crops were grown. 2Most critical period with respect to fruit yield except Van Doren greenhouse tomato work was vegetative growth. which 62 'temperature, oxygen activity in the soil solution and meteorological data. ‘Despite these limitations it is interesting to note the similarity in values for critical oxygen diffusion rates for a given plant species obtained by different investigators. It is quite possible that refine- ments in the accumulation of data for relating plant growth to soil aeration conditions as indicated in this thesis could bring universal agreement with respect to the effect of aeration conditions on plant growth. chin—.1 63 SUMMARY Quantitative study of factors affecting the oxygen diffusion current measurement with the platinum microelectrode has shown that: 1. Since the current is directly proportional to the length of the platinum wire, maintenance of a constant, known length of wire is essen- tial for accurate determination of oxygen diffusion rates. M a. ‘39:.I-1 o . 2. The exposure of copper oxide to the soil solution increases the observed.current. 3. Substances reducible at the potential used to reduce oxygen are not expected to be present in sufficient quantity in most mineral soils to appreciably influence the observed current. h. Provided the microelectrodes are not allowed to remain in the soil for periods of time longer than necessary to make the oxygen diffusion measurements and all other requirements are met, the reaction character- istics of the microelectrodes should remain constant and uniform for an entire summer of field use. S..A complete moisture film on the platinum surface is necessary for reliable determination of oxygen diffusion rates. 6. Oxygen diffusion currents increase 2.8 percent per Centigrade degree rise in temperature. 7. It is necessary to use a reference cell that will not become concentration polarized. 8. pH of the soil solution does not affect the oxygen diffusion current between a pH of h.2 and 8.0. 6h 9. An applied potential of -O.6S volt between the platinum micro- electrode and the reference cell is recommended for reduction of oxygen. 10. In most soils or soil conditions the electrical resistance of the soil has no effect on the 5, minute diffusion current. 11. Sufficient sampling accuracy in most soils may be obtained with 140 individual diffusion current determinations per sampling site without .w, expending an excessive amount of time. The effects of ionic strength of the soil solution and oxidation reduction potentials in the soil on oaqrgen diffusion currents have been studied, but the data are not complete enough to present at this time. Q It appears that pH, oxidation reduction potentials and electrical resist- ance of the diffusion medium are interrelated with the applied potential to the microelectrode, but their exact relationships and effects on the resulting electrical current are not as yet understood. Requirements for the study of the relationship between oxygen dif- fusion measurements with the platinum microelectrode and plant growth are discussed. The results of greenhouse and field studies are: 12. From an experiment in the greenhouse and another in the field the most critical portion of the growth period of tomatoes with respect to rate of oxygen supply appears to be during the early portion of the growing season. 13. Oxygen diffusion measurements with the platinum microelectrode correlated well with sugar beet yield in 1955, 1956 and 1957 on the Ferden rotation plots. 65 lb. Corn yields for these years were influenced to a much lesser degree than sugar beets, if at all, by changes in soil aeration conditions. 15. The possibility of estimating the influence of soil aeration conditions on crop yield with a knowledge of the changes in oxygen dif- fusion rates with time following rainfall or irrigation is introduced. The data presented in this thesis indicate the adaptability and ‘ E usefulness of the platinum microelectrode as a method of relating soil aeration conditions to their possible effects on plant growth. Problems requiring further investigation are the determination of critical oxygen fan-v.5."- arr-xii; "- .-'.- "" ~ diffusion rates as described, and the development of a rapid, accurate measurement of the concentration of oxygen in the soil atmosphere or soil solution. 66 BIBLIOGRAPHY l. Archibald, A. J., Effect of Soil Aeration on Germination and Develop- ment of Sugar Beets and Oats, M. S. Thesis, Michigan State College, 1952. 2. Bertrand, A. R. and Kohnke, H., Subsoil Conditions and Their Effects on Oxygen Supply and the Growth of Com Roots, Soil Sci. Soc. Amer. Proc. 21:13S-ho, 1957. 3. Cline, R. A., The Affect of Applied Fertilizer and Oxygen Diffusion Rate on the Growth, Yield and Chemical Composition of Peas, M. S. Thesis, Michigan State University, 1957. h. Dortignac, E. J ., Design and Operation of the Rocky Mountain Infiltro- meter, Rocky Mountain Forest and Range Experiment Station, Station Paper No. 5, February, 1951. 5. fiandbook of Chemistry and Physics, 35th Edition, Chemical Rubber Publishing 00., Cleveland, Ohio, l953-5h. 6. Hanks, R. J. and Thorpe, F. 0., Seedling Emergence of Wheat as Related to Soil Moisture Content, Bulk Density, Oxygen Diffusion Rate and Crust Strength, Soil Sci. Soc. Amer. Proc. 20:307-10, 1956. 7. Jackson, L. P., Effect of Soil Water Content and Oxygen Diffusion Rate on Growth of Potato Sets, M. S. Thesis, Michigan State University, 1956. 8. Kolthoff, I. M. and Lingane, J. J. , Polarography, Interscience Publishers, Inc., New York, 19141. 9. Laitinen, H. A. and Kolthoff, I. M., Voltammetry with Stationary Microelectrodes of Platinum Wire, Journ. Phys. Chem. h5le61, l9h1. 10. Lemon, E. R. , Soil Aeration and Its Characterization, Ph. D. Thesis, Michigan State College, 1952. 11. Lemon, E. R. and Erickson, A. E., The Measurement of Oxygen Diffusion in the Soil with a Platinum Microelectrode, Soil Sci. Soc. Amer. Proc. 16:160-63, 1952. 12. Lemon, E. R. and Erickson, A. E., Principle of the Platinum Micro- electrode as a Method of Characterizing Soil Aeration, Soil Sci. 79:383-92, 1955- 67 13. Meyer, B. S. and Anderson, D. B., Plant Physiology, 3rd Edition, D. Van Nostrand Co., Inc., New York, 1952. it. Raney, w’. A., Field Measurement of Oxygen Diffusion Through the Soil, Soil Sci. Soc. Amer. Proc. 1h:61-5, 1950. 15. Russell, M. B., Soil Physical Conditions and Plant Growth, Academic Press, Inc., New York, 1952. 16. Van Doren, Jr. , D. M., Field Adaptation of Oxygen Diffusion Measure- ments with the Platinum Microelectrode, M. S. Thesis, Michigan State University, 1955. 17. Van Doren, Jr., D. M., and Erickson, A. E., Construction of the Apparatus for Measuring Oxygen Diffusion with the Platinum Micro- electrode, Manuscript. 18. Van Doren, Jr., D. M., and Erickson, A. E., Modification of the Haney Oxygen Diffusion Tube, Manuscript. 19. Wiegand, C. L. and Lemon, E. R., A Field Study of Some Plant-Soil Relationships in Aeration, Manuscript. 20. Wiersma, D. and Mortland, M. M., Response of Sugar Beets to Peroxide Fertilization and Its Relationship to Oxygen Diffusion, Soil Sci. 21. Willard, H. H., Merritt, L. L. and Dean, J. A., Instrumental Methods of Analysis, D. Van Nostrand Co., Inc., New York, 1953. * __n - fl“ 2"; h - 2 " ’ “ 7,—1-3-m ., i F1 7 VITA Name: David Miller Van Doren, Jr. Born: Urbana, Illinois, August 16, 1932 Academic Career: l9h6-5O York Community High School Elmhurst, Illinois 195051; University of Illinois Urbana, Illinois 195h—58 Michigan State University East Lansing, Michigan Degrees Held: Bachelor of Science, 195k University of Illinois Master of Science, 1955 Michigan State University 68 [iii-ll, 1': ' F. L ””TITI'ITrfilfiufii‘fliMilfimll’imiimfl'“ 177 3421