Le magnesium d'Ubatuba et le ratio d'isotopes

Hill, Paul R.Hill, Paul R.: Unconventional Flying Objects, Hampton Road, 1995, USA, Section 16, 'UFO Artefacts', pp. 226-234, 1995

Une des histoires les plus cr�dibles de d�bris d'ovnis est celle concernant ce qui en est venu à s'appeler les fragments de magnesium d'Ubatuba - La cr�dibilit� �mane des faits accr�ditants d�velopp�s par les tests en laboratoire du gouvernement br�silien montrant l'�chantillon test� comme �tant constitu� à pratiquement 100 % de magnesium pur et totalement d�nu� des �l�ments m�talliques sous forme de traces, caract�ristiques d'une fabrication sur Terre. Un autre r�sultat �trange qui a simply baffled the laboratory scientists was the finding that, although pure, the magnesium was 6.7 percent heavier than ordinary pure magnesium. On reading the elegantly given account of this story in ''Flying Saucers, The Startling Evidence of the Invasion From Outer Space'' (Chapter 9) about a decade ago, I was impressed by the possibility that a non-universal isotope ratio might be the cause of the weight discrepancy. If so, it would also explain the purity from other elements. This would be strong evidence of extra-terrestrial manufacture because, even now, the only isotopes to have been separated on a significant scale are those of uranium 235 and 238. I ran through some trial computations. Sure enough, if the magnesium tested was pure isotope magnesium 26, mg26, its density would check the measured value to within about one fourth of one percent. This small discepancy could be experimental error, while the 6.7 percent couldn't be. I was amazed at this finding, and was even more surprised that the materials experts had not come out with it.

Confusion was added to fact when the third fragment, tested in the United States, turned out to be of a less impressive composition. But to understand this story, it is necessary to start at the beginning.

On September 14, 1957, the Rio newspaper ''O Globo'' published a report by the columnist Ibrahim Sued stating that there had been eye witnesses to a UFO explosion over the sea near Ubatuba, Sau Palo, Brazil. The eye witnesses were on the beach when they saw a ''flying disc'' approach at fantastic speed and pull up to avoid striking the water. Still traveling rapidly, it exploded into thousands of fiery fragments which fell into the sea. A number of small fragments fell near the beach and some were recovered. A witness sent 3 small fragments in a letter to Mr. Sued. The capable and productive Brazilian UFO investigator Dr. Olavo T. Fontes, in a fast bit of footwork, interviewed Mr. Sued on the day the story broke. Letting the columnist know he had connections for the scientific analysis of the samples, Fontes walked out with them.

The three fragments, weighing a few grams each, were gray in color and had very irregular surfaces. Their appearance suggested that they were fragments of a much larger piece or object. The surfaces were covered in scattered areas with a thin layer of a white powdery substance, adherent to the surface. They appeared burned with fire or heat, and had many small cracks in the surfaces. The fragments were numbered 1, 2, and 3, and then fragment 1 was cut into a number of subpieces for testing. All the irregular surfaces were cut away, leaving little pieces with polished surfaces weighing about 0.6 gm each.

Subpieces of sample 1 were analyzed by the Mineral Production Laboratory, a division of the National Department of Mineral Production in the Agricultural Ministry of Brazil. This laboratory is the official Brazilian laboratory for the analysis of minerals. The analysis was done under the direction of the laboratory's chief chemist, Dr. Feigl from Germany. Following a simple test on one of the pieces to prove the sample a metal, he decided on a spectrographic analysis to determine the composition. Each metallic element has a characteristic set of spectral lines which are uniquely its own and the tests, while destructive, can be extremely sensitive.

The first spectrographic analysis was made by Dr. Louisa Maria A. Barbosa. One of the pieces was burned in an arc in a routine test to show the metal was magnesium. Then a second piece was burned in a large Hilger Spectrograph to determine the metallic impurities (trace elements) present. Surprisingly, there was no trace of any other metallic element - a very spectacularly pure magnesium. A test by another operator confirmed this finding. However, six faint lines not belonging to any metal showed the presence of a mere trace of a non-metallic substance.

Since the magnesium purity was so unusual, even unheard of, it was decided that an x-ray crystallography test should be made both as a confirmation and to determine the non-metallic constituent. Dr. Elysiario Taravora Filho, head of the Laboratory of Crystallography, carried out the sophisticated x-ray diffraction tests needed. The results again showed pure metallic magnesium with a faint trace of the hydroxyl radical, as in magnesium hydroxide. The x-ray diffraction tests not only confirmed the purity of the magnesium, but clearly explained the nature of the white powder on the fragment surfaces as being magnesium hydroxide formed when the pieces hit the water. They also showed that the trace of hydroxide was something probably not present before the material hit the water. To complete the picture, it was clear that the fragments had dropped into the water in an incandescent state as the witnesses said, because only at elevated temperatures could the hydroxide have migrated to the interior of the samples. Finally, the six faint lines previously unidentified on the spectrographs were checked out as being due to the slight trace of magnesium hydroxide.

It only remained to check the material density in gm/cc, taking the density of pure water at 4 degrees centigrade as unity. The density of fragment 1 was determined for a small, carefully polished piece taken from teh center of the fragment. The standard procedure of suspending the chip from a Jolly balance and taking the ratio of the weight when suspended in water was followed. Operators and spectators alike were in for a big surprise. Three successive tests all showed a density (specific gravity, if you prefer) of 1.866. The expected density of magnesium was 1.741 !

Repeated measurements in agreement showed accidental errors to be negligible. However, systematic errors are doubtless present and are difficult to analyze. No mention was made of the corrections needed or made for the buoyancy of the fine wire generally used to suspend the sample in such a test, nor of the effects of water surface tension on the wire. Such systematic errors and their calculated corrections might account for a fraction of one percent error in the density, but by no stretch of the imagination for the difference between 1.866 and 1.741. Nor could the minute trace of hydroxide account for it.

Everyone was baffled. Fontes discussed isotope ratiaos and the possibility of their great importance. He also considered the possibility of a previously unknown closely packed crystalline structure. Of the three possible explanations, he concluded:

  1. The crystallography tests showed standard hexagonal close- packed crystals.
  2. Hydroxide was present in too small an amount to account for the discrepancy.
  3. The density ratio gave no ground for reliance on an unusual isotope ratio.

Fontes felt that a program to search for a low-density sample should be undertaken, but it wasn't. The situation was left as a baffling mystery. Fontes felt well satified by the extreme purity result in itself; at least his words seem to say so.

I totally disagreed with the third conclusion. The following elementary computations are based on the fact that the volume of a given number of atoms of magnesium is fixed by the outer electron shells, which is the same for each isotope, within close limits; hence the density is proportional to the isotopic weight. The (equivalent) atomic weight of magnesium is computed in the table:

Table XVI-1. Computation Table
(1) Isotope (2) Abondance relative (3) Poids atomique (4)=(2)x(3) Poids relatif
Mg24 0.786 23.99189 18.85736
Mg25 0.101 24.99277 2.52427
Mg26 0.113 25.99062 2.93694
1.000 Sum = 24.31884
= atomic weight magnesium

Then the proportion is: The density of a pure isotope of magnesium is to the normal density of magnesium as the atomic weight of the isotope is to the (normal) atomic weight of magnesium. We thus obtain the desity of pure mg26:

Atomic Wt Mg26 Density pure Mg26 = --------------- x normal density Mg Atomic Wt Mg

= (25.99062 / 24.31884) x 1.741

= 1.861 g/cm3 to four figures

The experimentally measured value of fragment 1, 1.866, is only 0.005 higher than the theoretical value for pure magnesium 26 ! This amounts to 0.005/1.861 = 0.27 percent, say a quarter of one percent higher than theory. As already mentioned, the most probable source of such a small discrepancy is the procedure for weighing a very small specimen submerged in water, plus the inclusion in the test of foreign material, such as the trace of migrated magnesium hydroxide.

These computations give a strong indication that fragment 1 was the pure isotope magnesium 26. This not only solves what is otherwise a density mystery, but explains the phenomenal purity at the same time, for there is no surer way to obtain purity than by the separation of isotopes. Still, it is just an indication, not a measurement. Fragment 1 absolutely should have been tested in a mass spectrometer. This gives the last word on isotope ratios and on chemical contaminants present as well.

As soon as the magnesium of fragment 1 was determined to be of rare purity, the interested parties should have reacted like diamond merchants with three rare diamonds to cut. Instead they made a double error: they didn't test fragments 2 and 3 at all, and they tested fragment 1 to extinction.

There was no ground for assuming that the fragments were basically identical, for they probably came from different parts of the disk. They might have had totally different uses and been of different compositions to suit. Fontes' discussion indicates that the laboratory personal, Professor Filho in particular, knew the full implications of finding a specimen of such rare purity, a ''Star of India'' among metals. Didn't they want to know if fragments 2 and 3 were crown jewels also ? Fontes had given one subpiece of fragment 1 to the Brazilian Army at their request, and another to the Brazilian Navy at their request. The last of the subpieces went to Prof. Filho, who powdered them one by one as he repeated his pet diffraction tests to measure the magnesium hydroxide content. These relatively unimportant tests could have been accomplished with subpieces of fragments 2 and 3 as they all contained the hydroxide from the reaction with sea water. Had he done so, he would have had a preliminary reading on the composition of the other two fragments. Instead, the last pieces of fragment 1 just went down the drain. One or two of these should have gone to the United States or Europe for tests in a mass spectrometer. One or two should have been preserved in a bottle of dry helium and offered to major world museums as relics.

According to ''UFO, The Whole Story'' (Lorenzen 213), Dr. Fontes sent the remaining fragments to APRO to be tested in the United States. Public record is a bit nebulous as to who tested which fragment. Oak Ridge and Dow Chemical both tested subpieces of one fragment. The only highly unusual result to be reported from this pair of tests was that the fragment contained possibly as much as a thousand parts per million of aluminum. As can be seen in Table XVI-2, aluminum in such a quantity is not expected in a commercially produced pure magnesium. A fragment subpiece was also given to the U.S. Air Force even though they would not see to the presence of an APRO representative during the tests. When the Air Force reported that they had burned up the sample in a spectrometer test without result, and asked for more, naturally Mrs. Lorenzen refused.

The remaining fragment was turned over to the University of Colorado UFO project. This fragment weighed just under 5 grams and was about the size of the last joint of your little finger (Saunders and Harkins). Since it was an irregular piece, it must have been the original fragment 3, as Dr. Fontes reported fragment 2 was cut and polished before it was sent to APRO. Appropriately, the project's physical chemist, Dr. Roy Craig, assumed responsibility for the test program. He decided on a neutron-activation procedure which can be effective yet economical of material. In this procedure, the subsample to be tested is placed in the center of an atomic reactor where it is radiated with neutrons, making the material radioactive. Removed from the pile, the sample is placed in a gamma-ray spectrometer which can identify the various elements to within a few parts per millio. More detail of the test procedure is given in Chapter 3, written by Dr. Craig, of 'UFOs ? Yes !' The tests were conducted by the FBI's Alcohol Tax Div. Laboratory in Washington and were personally witnessed by Dr. Craig.

Table XVI-2. Elements Found in Comparative Magnesium Samples (Concentration in parts per million)
El�ment Ovni br�silien Magnesium Dow
Manganese 35 ppm +/- 14 percent 4.8 ppm +/- 10 %
Aluminium not detected not detected
Zinc 500 ppm +/- 20 percent 5.0 ppm +/- 20 %
Mercury not detected 2.6 ppm +/- 19 %
Chromium 32 ppm +/- 31 percent 5.9 ppm +/- 20 %
Copper 3.3 ppm +/- 30 percent 0.4 ppm +/- 50 %
Barium 160 ppm +/- 12 percent not detected
Strontium 500 ppm +/- 20 percent not detected

Only a small chip of the main sample was used, and it was tested for the presence of the eight metals listed in Table XVI-2. The quantity of each was determined in parts per million to the percentage accuracy shown. The middle column gives the results for the Brazilian UFO, and the right-hand column the corresponding results for Dow commercially pure magnesium.

Dr. James A. Harder of the University of California Civil Engineering Department (Symposium on Unidentified Flying Objects) makes the additional point that no trace of silicon or calcium was disclosed by the tests, either. He also adds that, of all the elements, calcium is the most difficult to remove when a serious attempt is made to achieve pure magnesium by earth technologies.

In view of the absence of elements difficult to control, and the presence in greater amount of elements easily removed if desired, such as barium and strontium, one can gather the impression that the composition is intentional, well controlled, and unique. In the official report, Dr. Craig said, ''Although the Brazilian fragment proved not to be pure, as claimed, the possibility remained that the material was unique. The high content of strontium was particularly interesting, because strontium is not an expected impurity in magnesium made by usual production methods.'' Need more be said ?

In ''UFO, The Whole Story'', the Lorenzens made one of the most significant comments about the Ubatuba magnesium fragments: ''The really strange facet of this particular incident and the subsequent investigations is the fact that three different analyses performed on three different fragments yielded completely different results.'' Agreed ! Yet all of the analysts are probably correct in their physical findings.

Anyone experienced in missile development can explain the quoted result. When a supersonic developmental missile fails by disintegration it usually breaks into what looks like hundreds of fragments which seem to fill the sky like confetti. All parts of the vehicle get scrambled together. On reaching Earth they are better shuffled than a deck of cards on the deal. Any complex Earth machine is composed of dozens of metallic alloys, a different alloy for every special purpose. UFOs may well use special purpose materials also. The point is that the differences may be intentional. David Saunders also makes this point.

Thus the best overall view of the Ubatuba magnesium fragments is that each one is unique to its special purpose. Fragment 1 was the most intriguing of them all, for according to the tests of some of the world's most experienced metallurgists that fragment was absolutely pure of all metals other than magnesium. This is as would be expected if it were the pure isotope magnesium 26. Fragment 1 was also the only fragment with the density to be expected of this pure isotope.