Free Science Fair Supplies Search! Enter Your Topic Below:

4. CHEMISTRY

1

In the course of developing a process for the production of magnesium, James C. Hartman of Fort Wayne, Indiana, became interested in the great possibilities of titanium. Learn­ing that the high cost of producing titanium was due to the complicated decomposition process, he decided to investigate the possibilities of producing titanium by a cheaper method. His project, "Electrolytic Decomposition of Titanium Diox­ide," was exhibited at the Ninth National Science Fair.

"Discussion

Several excellent books on the subject were procured from our public library and a great deal of valuable information from current magazines, especially from Modern Metals. Literature from the Dominum Magnesium Company in Canada proved a great help by explaining the process com­monly used to produce titanium and by pointing out that titanium is one of the few metals that will combine directly with nitrogen from the air, thus making it necessary for the reaction to take place in an inert gas atmosphere. The report indicated, too, that the future of titanium production lay in the development of an electrolytic decomposition process. With this in mind, I began working on the development of a process that would involve electrolytic decomposition. Since titanium dioxide is the compound currently used for the extraction of titanium, the first big problem was to find a suitable solvent for titanium dioxide. This solvent would have to melt at fairly low temperatures and be able to dissolve large amounts of titanium dioxide. I first tried a nitrate, be­cause of the low melting point of most nitrates. In order to keep a uniform heat and effect the decomposition of the nitrate, a porcelain heater cone was used. With this, several nitrates were melted (magnesium nitrate, aluminum nitrate, potassium nitrate and strontium nitrate) but they did not dissolve an appreciable amount of titanium dioxide and the nitrates began to decompose when heated for any length of time.

After several unsuccessful attempts with the nitrate baths, I decided to abandon the electrolytic method of decomposi­tion and try decomposing the titanium dioxide by passing it into an alcohol vapor chamber and applying heat. This ex­periment was set up according to the diagram in Fig. 1 of the appendix. The reaction intended was 2CH3OH-j-3TiO2→2CO2 + 4H2O + 3Ti but this reaction required a higher temperature than was practical. Eager to exhaust the pos­sibilities of this method, I repeated this experiment, this time using a pyrex tube and an acetylene torch for heat. Instead of the alcohol vapor, producer's gas, acetylene propane and carbon monoxide were used this time. The pyrex tube melted before any decomposition of titanium dioxide occurred. Sat­isfied that this type of decomposition did not hold promise, I resumed the original plan and the search for a suitable solvent.

Further reviewing and reading brought to light a mineral called sphene, a mixture of calcium oxide, titanium dioxide and silicon dioxide. This seemed to point up the possibility of using glass as a solvent, and the fact that glass melted at a fairly low temperature gave further encouragement along this line. Having ground small pieces of glass into a fine powder, I added an equal amount of titanium dioxide. This mixture was then heated in a crucible, and the titanium dioxide dissolved very readily even when additional portions of the compound were added.

Electrolytic   Decomposition               

The electrolytic method for the decomposition of titanium dioxide consists of dissolving titanium dioxide in a mineral called sphene. Sphene is a mixture of titanium dioxide plus calcium oxide plus silicon dioxide. This mineral was produced artificially by crushing scrap transparent glass, which is a mixture of calcium silicate and silicon dioxide, and adding titanium dioxide to this. The mineral formed when the glass melted and dissolved the titanium dioxide. More titanium dioxide was added when the sphene was melted, so that there was an abnormal amount of titanium dioxide in the sphene. The temperature maintained in the cell was ap­proximately 1700-1800° C. The glass will melt at approxi­mately 1650-1710° C. This heat was sufficient for the reaction to take place, but since titanium will not melt until 1810° C, it would be more profitable to maintain a temperature in an excess of 1810° C. The titanium metal could then be tapped off periodically in a liquid state. The nature of the electrolyte also permits the addition of more titanium dioxide at any time. The metal as found in the cell was in the form of very fine powder at the negative pole, which is the wall of the cell. The metal was attached to the electrolyte, which made it necessary to leach it out with hydrochloric acid. Titanium metal is highly resistant to dilute hydrochloric acid, but the sphene is slowly attacked by it.

I rented 5 6-volt storage batteries and hooked them in multiples to supply the necessary electricity. The 5 bat­teries supplied about 500 amps., but the resistance in the cell and the connections greatly reduced this amount. The third time the process was tried a 7-volt, 500-amp. plating generator was used. The decomposition can begin as soon as the electrolyte is molten. The electrical resistance provided by the electrolyte is great enough to keep the electrolyte molten for a while after the external heat is turned off. The external heat was supplied by a 1,000,000 B.T.U. burner at the Sturgis Light Metals Foundry.

The specific gravity of sphene is 3.4 to 3.6, while the specific gravity of titanium is 4.5. The difference between the two specific gravities is great enough to allow a definite separation between the titanium and the electrolyte. The titanium will go to the bottom of the cell unless the cell is violently agitated.

The cell consists of a nipple three inches in diameter and four inches long, with a pipe cap on each end. A hole in one end allows the graphite (+) electrode and an asbestos diaphragm to enter the cell. The diaphragm is needed because the oxygen is readily dissolved in the glass as well as the electrolyte. The negative pole of the cell is the metal part (eg., the two pipe caps and the nipple).

The first cell did not have a carbon lining in it and was severely attacked by the electrolyte and combined with the titanium to form ferrotitanates. The idea of using a carbon lining came from the one used to protect the cells in alumi­num production. The lining is made by mixing lampblack with tar and coating the inside of the cell with it. The cell is then heated to about 700° C. to drive off the volatile material in the tar, leaving the carbon.

The efficiency of the cell is difficult to determine when worked on a small scale, but it does seem to be considerably less  expensive to  operate  than  the  normal  decomposition process. The high temperature required and the resistance of the electrolyte tend to lower the amperage greatly. The anode also is attacked by the electrolytic decomposition of titanium dioxide more than by aluminum oxide. For every atom of titanium liberated, two atoms of oxygen are liber­ated; while for every two atoms of aluminum liberated, three atoms of oxygen are liberated. However, the process shows possibilities of overcoming these smaller handicaps with the greater advantages. The process is continuous in that the cell metal collects on the bottom of the cell and can be tapped off periodically. The titanium dioxide also can be added periodically to keep the process in operation. Also, in contrast to the conventional decomposition method, a less noble metal like magnesium or sodium is not needed for the decomposition. The only cost in producing the electrolyte is in crushing the glass and the cost of the titanium dioxide itself (rulite ore is about 6 cents a pound). The titanium would not have to be converted to titanium tetrachloride, thus saving the cost of chlorination.

Bibliography

Barksdale, Jelks; Titanium. New York: Ronald Press Co. 1949.

Bray, John L.; N on ferrous Production Metallurgy. New York: John Wiley & Sons. 1947.

Conant, James Bryant and Blatt, Albert Harold; Chemistry of Organic Compounds. New York: Macmillan Co. 1952.

Everhart, John Laurence; Titanium and Titanium Alloys. New York: Reinhold Publishing Corp.  1954.

Graham, A. Kenneth; Electroplating Engineering Handbook. New York: Reinhold Publishing Corp. 1955.

Liddell, Donald M.; Handbook of Nonferrous Metallurgy. New York: McGraw-Hill Book Co. 1945.

Longe, Norbert Adolph; Handbook of Chemistry. Sandusky, Ohio:  Handbook Publishers, Inc.  1956.

Longsdorf, Alexander S.; Principles of Direct-Current Ma­chines. New York: McGraw-Hill Book Co. 1940.

Schwarzkopf, Paul; Powder Metallurgy. New York: Mac­millan Co. 1947.

Timm, H. A.; "The Potential Merits of Titanium and Tita­nium Alloys Produced by the Oxide Reduction Process." Presented at the A.I.M.E. Regional Reactive Metals Con­ference in Buffalo, New York, 1956."

2

When he was awarded the unusual opportunity of working in the University of California Radiation Laboratory during the summer of 1957, Neil Nininger of Larkspur, California, worked on the problem of how to make tantalum carbide filaments that would not develop localized hot spots and burn out prematurely. The TaC filaments, which theoretically should burn at 3800° C. for short periods, were to be used in the radiation laboratory mass spectrographs. Neil's in­vestigation was so successful and was considered so important that he was invited to make a report on his work to the personnel of the entire laboratory. He submitted a report of his work, which he called "The Production of High Temp­erature Tantalum Carbide Filaments," as part of his entry in the Science Talent Search the following fall. He was judged one of the winners and was awarded one of the top scholar­ships. Section III of bis report outlines the experimental results of his work.

'TaC, Ta2O and Ta all have different crystal types, and the junctions of the various phases are poor electrical con­nections. Consequently, hot spots develop at these junctions of high resistance, burning out the filament. Thus, the prob­lem becomes how to produce filaments whose phase junc­tions1 are increased in area. When this is done the total resistance across the junction is lowered.

The effect of temperature on reaction rates is such that when a filament is run at a low heat (800° C), the reaction is slowed to such an extent that the formation of the various phases can be observed. Under this condition, Ta2C first forms in the hottest part of the filament, the region noted as "A" in Figure 2. As the Ta2C has a higher resistivity than the Ta metal, the "A" region becomes hotter and begins to extend itself over the length of the filament. At the same time, TaC forms in the hottest part of the "A" region, at "B." The TaC formed has a lower resistivity than the surrounding Ta2C, and forms a cold spot. The cold region of TaC extends itself along the filament, toward the Ta-Ta2C junction. In a short time, the TaC reaches the Ta-Ta2C junction and a very localized transition zone is formed. The resulting transition zone (Figure 3) has a high resistivity because of the low cross-sectional area of contact, and the poor electrical contact between different crystal types. The filaments burn out at this point.

1. There must always be junctions, because the extremities of the filament are cooled below the reaction temperature by end con­duction of the support.

To make the area of contact gradual, I tried extending the transition zone over a larger length of the filament by carburizing rapidly. I flashed several filaments for short periods of time, at high temperature. This was to establish the Ta2C crystal structure along the "table" of the filaments, and well down the "legs" (Figure 2). I then ran the filaments for long periods of time at low temperatures (800° C), to form TaC over the Ta2C crystal structure. But the transition zones were again too localized, and the filaments burned out. I tried the above procedure again, except that after flashing the filament I ran it for long periods of time at higher temperatures (1400° C). However, the transition zones, although pushed down to the legs, became localized again, causing the filament to burn out.

After failing in this method of attack, I reinforced the legs of the filaments ("C" in Figure 2) with .004" Ta foil strips. When the transition zones are pushed down to this reinforced region, they have a larger cross-sectional area with lower resistance. The reinforced transition zones also have higher mechanical strength. These factors enable fil­aments made by this last method to be used in mass spec- trographs."

3

An interesting project that involved both construction and research was done by Leaf Turner, Brooklyn, New York, a member of the Honors Group of the Eighteenth Science Talent Search. The title of his project was "Hydrogen-Oxygen Fuel Cell."

"Recently, a device that has been known since the begin­ning of the century has finally received its deserved recogni­tion. This is the fuel cell. The significance of the fuel cell is that electricity is generated directly from fuel by an electrochemical reaction.

Because of the importance that might result from further research concerning this cell, I decided last June to collect as much data about it as I could. Unfortunately, there was relatively little information. In the July General Electric Review, I came across an article on the fuel cell. I wrote to the authors, Dr. Liebhafsky and Dr. Douglas, requesting further information. They generously forwarded me enough for me to begin my work.

In July, also, in the Journal of the Electrochemical Society, I found "The Fuel Cell Round Table," which contained a summary of all the work that had been done on the fuel cell.

There were three kinds of fuel cells suitable for my project: a redox cell, a hydrogen-oxygen fuel cell, and a high temperature carbon monoxide-oxygen or hydrogen-oxygen cell. I decided that the hydrogen-oxygen cell was best suited for construction and laboratory experimentation.
Construction of Cell

The two porous graphite electrodes that I am using were obtained from two dry cells. I feared that the not too delicate strokes of the hammer necessary to break open the dry cells might cause the electrodes to break and that much time would therefore be lost. Fortunately, this was not the case.

At this point, my problem was to devise an efficient method of having oxygen adsorbed onto the surface of one electrode and hydrogen adsorbed onto the surface of the other. I cut off from the electrodes the ends having the metal binding posts. I obtained the use of a lathe with which I drilled a hole four inches into the electrodes from one of the ends. These holes had a diameter 1/8 inch greater than the diam­eter of the glass tubing, 3/16 inch, which I later inserted into these holes for the gas inlets. Half of my problem was solved, since the gas now would pass very close to the surface of the electrodes.

The other half of my problem was to obtain proper cata­lysts to adsorb these gases. With the help that General Electric offered me, I was able to choose my catalysts. For the hydrogen electrode I decided to use platinum from thermally decomposed platinic chloride; for the oxygen electrode I adopted the recommendation of Kordesch and Marko, which was a solution of 2.4 g Co(NO3)2 • 6H2O plus 6.2 g. Al (NO3)3 • 9H2O in 100 cc. of water.

In order to coat the surface of the hydrogen electrode, I poured a solution of platinic chloride into the center of the electrode. I heated the electrode until the water evaporated and the surface was coated with platinic chloride. Then I carefully heated the entire electrode until I saw a metallic coat on the electrode which I knew was thermally decom­posed platinum.

In the other electrode, I evaporated 80 cc. of the nitrate solution and heated the entire electrode at 800° C. to convert the nitrates to oxides. I have had to take great care in the heating of these electrodes for fear that they might crack.

My next step was to find some type of cap to cover the openings of the holes, through which I could pass two tubes, one an inlet for the gas and the other an outlet. I decided to drill a three-quarter-inch hole into two one-holed rubber stoppers in order to cap the electrodes. This hole is smaller than the diameter of the electrodes, so that the stoppers might fit airtight over the electrodes. Then, through the side of the stoppers, I drilled a 3/16-inch hole to fit the outlets. After I put the stoppers over the electrodes and inserted into the holes the glass tubing for the inlets and outlets, I connected the inlets with an air supply. I found that a great amount of air was leaking from the area between the stoppers and the electrodes. In order to correct the leak, I used rubber cement. I found this unsatisfactory because the air, which was under considerable pressure, bubbled through the cement. Next I tried Duco cement. This seems to be serving my purpose well up to the present. I anticipate, however, that even the cement will not be completely satis­factory because it is inflammable. If necessary, I may try Miracle adhesive; but, here again, I may have the same problem.

At this point, I had to obtain a three-neck flask, in which I could put my electrolyte (KOH solution) and into which I could insert the two electrodes, and a Liebig condenser, which would condense any water that might evaporate from the electrolyte. I could not fit the electrodes into the necks of the flask I obtained, because the diameter of the necks was slightly too small. I decided to sand down the electrodes.

In the next phase of the experiment, I am going to fit the two electrodes into the two necks of the flask and a Liebig condenser into the third. My electrolyte will be a solution of 30% potassium hydroxide in water. Then I shall have to obtain my hydrogen supply.

Theory of Fuel Cell

The principle of this device is as follows. Oxygen molecules are adsorbed on the surface of the positive electrode. This adsorbed layer, being more active than molecular oxygen and being free, combines with the water in the electrolyte to form two hydroxyl ions. Each of these ions, having a single negative charge, removes an electron from the oxygen electrode, making the electrode positive. These hydroxyl ions migrate to the other electrode and combine with the adsorbed hydrogen to form water, depositing electrons in the process. Since the hydrogen electrode is negatively charged and the oxygen electrode is positively charged, electrons may flow through the external circuit. The following equations sum­marize the foregoing.

Hydrogen Electrode (Negative Pole):

H2 = 2H

2H + 2OH- = 2H2O + 2e

Oxygen Electrode (Positive Pole):      

½O2 = O

O + H2O + 2e = 2OH-

Over-all Cell Reaction:

H2 + ½O2 = H2O

After I have my fuel cell working, I shall plot its various characteristics, e.g., current density versus voltage at different temperatures. Basing my information on what I have read on the low temperature hydrogen-oxygen fuel cell, I should obtain approximately .73 volts and a current density of 1 mA/cm2.

Advantages of the Fuel Cell

As the world population increases, demand for electricity will increase. Fuel resources are constantly diminishing and the cost is rising. A more efficient and economical method of producing electricity must be devised.

The present method of producing electricity involves the use of steam engines. The efficiencies that have been obtained are about 30%. Thermodynamic laws set an upper limit on the efficiency.

The fuel cell, however, converts all chemical energy directly into electricity without the loss of energy incident to the use of steam engines.

Bibliography

1. Journal of the Electrochemical Society - Vol. 105, No. 7,
   July, 1958, pp. 428-431.
   
2. General Electric Review - July, 1958.
   
3. Fifth  World   Power   Conference  - Recent Research  in
   Great Britain on Fuel Cells.
   
4. Bauer - OTS Report - Fuel Cells."
   

4

An Honors Group Project of the Eighteenth Science Talent Search, entitled "Ozone and Sodium Hypochlorite," was done by Burton J. Krohn of Nashville, Tennessee.

"My project actually began with the building of a crude ozone generator as a chemistry project a year ago. The uniqueness of the chemical activity of this gas caused my interest in this subject to develop, and I decided to extend my work and research deeper, after further encouragement by an extra chemistry course last summer. The purpose of this project, as well as studying the effect of ozone on a hypochlorite bleach solution and quantitatively analyzing a solution of possibly four similar components, plus a gas, is to develop accurate and efficient scientific laboratory tech­niques.

The Ozone Generator

My ozone generator, simple but effective, consists of a 6-mm. pyrex glass tube bent into a right angle about 8 inches from one end and drawn into a jet about 10 inches in length at the other. The inner electrode is a 6-inch rat-tail file, and the outer electrode is aluminum foil wrapping. A 12,000-volt current produces a brush discharge around the glass. The flow of oxygen through the system is regulated by a homemade siphon mechanism and an adjustable pinch-cock, since an oxygen tank regulator costs over thirty dol­lars. The jet serves to enlarge contact surface area between gas and liquid as the oxygen-ozone mixture is bubbled through test solutions.

With moderate but growing success, I have kept the oxygen flow rate at 100 ml. per minute. Also to increase efficiency, water vapor, which catalyzes decomposition of ozone, is at least partially removed with calcium chloride. Despite precautions, the efficiency has constantly been ham­pered by leaks, which only recently I have succeeded in minimizing. The concentration of the ozone produced is approximately  1%.

Plan of Action

At the beginning of my project the question in my mind was whether ozone is sufficiently powerful to oxidize the chloride ion from the — 1 oxidation state to +1 in hypo­chlorite, +3 in chlorite or +5 in chlorate. This question was answered by H. Willard and L. Merritt, Jr., in the 1942 Analytical Edition of Industrial and Engineering Chemistry. They said the change effected by ozone upon the chloride ion is negligible. I then turned to the possibility of raising hypochlorite from +1 to a higher oxidation state. I found no information about such a reaction; therefore I chose this question as a project.

In the reaction between ozone and a hypochlorite there is a possibility of five resultant products:  unreacted CIO-, C1-, C1O2 C1O3 and the gas C1O2. The problem is the quantitative determination of each component and the estab­lishment of proportions in which they occur. After con­siderable research into chemical properties of these compounds and possibilities (including ion exchange resins) for their separation and analysis, I decided upon the following steps (see equations):

Before Ozonation

1. Iodometric determination of total active oxygen content.
   
2. Argentometric determination of total chlorine content. C1O" is reduced  to Cl" with hydrogen peroxide. This step accounts for any Cl" formed by natural decomposition of C1O".

After   Ozonation

3. Iodometric determination, in moles, of increase or de­ crease in total active-oxygen content. This test will not in-­ clude C1O3 if it is formed.
   
4. Argentometric  determination of total chlorine content C102 and C1O are reduced as in step 2 before ozonation. If total C1 is less than before ozonation, it may be due to one of two factors, or both. One possibility is the formation of CIO3, which, as I proved by experiment, is not reduced appreciably by H2O2, even when the mixture is boiled. If C1O3" is formed, it can be quantitatively reduced to Cl" by boiling with sodium nitrite. The other possibility is the loss of chlorine as C1O2 gas, whose quantity can be calculated by subtracting the total Cl content after all reductions, from the total Cl content before ozonation.

Chemistry

Separation of C1O" and C1O2 The mixture is titrated against a slightly basic solution of arsenite, with which only C1O" reacts, and quantitatively.  Methyl orange solution  is used as indicator.

1. Estimation of Cl" formed during ozonation. If the total chlorine found in step 2 after ozonation is greater than the sum of the chlorine in all C1O-, ClO2 and ClO3 determined after ozonation,  then this  difference resulted from one of two causes, or both. One possibility is the chloride ion formed during ozonation; the other is the remaining original chloride present before ozonation, due to previous decomposition of C1O". The chloride formed during ozonation may be calculated by subtracting this remaining chloride, which is the difference in moles between steps 2 and 1 before ozonation, plus the combined number of moles of C1O-, C1O2 and C1O3, from the total chlorine found in step 3.
   

Results

As of now I am well into step 1 after ozonation, and I have also begun work on more difficult step 2. In each test in the first step 2000 cc. of O2—O3 mixture is bubbled through 30 ml. bleach solution diluted to 1/100 of its original strength. Fifty ml. of 0.02 M potassium iodide solution, plus 1 ml. of 99% acetic acid, is used in the titration flask, and a standard 0.005 M solution of sodium thiosulfate is titrated against the liberated iodine. My starch indicator suspension (as recommended in "Iodometric Determination of Inorganic Substances," Volumetric Analysis, III) was prepared by making a paste of 2 g. cornstarch plus 1 mg. mercuric iodide as a preservative in 30 ml. distilled water, and adding this to sufficient boiling water to make 1 liter. I found this preparation to be very satisfactory in providing precise, clear end-points.

Recorded are results of twenty-seven tests (see table), which vary so widely that their average cannot be consid­ered conclusive. However, without exception, in each test the total active oxygen after ozonation is less than in the tests taken each day before ozonation, but the average number of moles lost is equal to only 3.2-4.7% of the amount of ozone used. Perhaps after tabulation of about sixty tests a more accurate figure may be obtained.

In my work in step 2, I add 2-3 ml. of full-strength hydrogen peroxide to 30 ml. test solutions, and boil for 5 minutes. The Cl formed is titrated against 0.01 M silver nitrate, using precipitation of red silver chromate as in­dicator. The major difficulty is distinguishing the end-point, which is not as clear as the starch indicator in iodometry. Ten inconsistent tests have shown my need for more practice in this titration.

Steps 3 and 4 remain in the future until my other data show conclusive results, and my familiarity with, and habits in, the chemical laboratory further improve.

I believe that this project in its planning, experimenting and calculating has thus far helped to develop, through failure and success, my persistence and attitude as a scientist.

Are You Ready To Move Onto The Next Lesson? Click Here….