·Table of Contents
·Materials Characterization and testing
Photoluminescence of synthetic and natural Materials for Industry and Handicraft
D. Ajò,F. De Zuane, L. Maini, G. Pozza,
ICTIMA, CNR - C.so Stati Uniti, 4 - 35127 PADOVA (Italy)
CSSRCC, CNR - Via F. Marzolo, 1 - 35131 PADOVA (Italy)
Dip. di Mineralogia e Petrologia, Università degli Studi - Corso Garibaldi 37 - 35137 PADOVA (Italy)
The photoluminescence (PL) of many materials is due to the presence of luminescent centers, such as first-series transition metal and rare earth ions. They can occur as natural impurities, or they may be introduced as intentional dopants.
The PL behaviour depends on the oxidation state of a dopant ion and on its coordinative environment. Within a class of similar objects we may detect slight chemical or structural differences, arising from the material history, from the mining of raw materials or the synthesis of the precursors, through production processes and subsequent treatments, until the making of the final products.
In the following sections we will show some applications of PL and the problems concerning the synthesis and the use of suitable reference samples for the PL analysis in different fields.
The PL spectra have been recorded at ICTIMA (CNR, Padova) by means of a custom-made apparatus (Fig.1), including several detectors suitable to each investigated spectral range and of a helium-flux cryostat, which allows measurements of the PL of several crystalline or amorphous samples at the same time, from liquid-helium to room temperature.
Fig 1: schematic view of the PL apparatus.|
For the excitation, one of the following continuous-wave laser sources has been employed: He-Ne (l=633 nm), doubled-frequency neodymium (532 nm), argon-ion (488 nm) and doubled-frequency argon-ion (244 nm). The laser beam passes through the hole H of a parabolic mirror (Fig. 1); the PL emitted by the sample S is reflected by the mirror, focused by the lens L and sent to a high-resolution grating spectrometer.
This apparatus is equipped with a versatile sample-holder, designed for the investigation of samples with any shape (even irregular), within a large range of size (from a few hundredths of a millimeter to several centimeters), as it can be seen in Fig. 2.
Moreover, the sample holder can be moved along the xyz directions, keeping untouched the optical alignment. Raw materials (as a sapphire in its mother rock  or a plant leaf ) and finished objects such as a Sumerian seal  or an artistic goblet  can be studied without any preliminary treatment.
In the case of heterogeneous materials, a mapping can be carried out by repeating the measurement in different points of the sample, with a resolution of a few tenths of a millimeter.
Fig 2:PL "heavy" apparatus during the analysis of a bottle's glass.
Fig 3: the new spectrometer equipped with the CCD detector.|
The spectra discussed hereafter were collected at room temperature, unless differently specified. The time of a single data collection by the above described system ranges from a few minutes to about half an hour, according to the required accuracy.
Moreover, the apparatus is mounted on a large and heavy optical bench. Therefore, we developed and are presently testing an alternative low-cost system based on a spectrometer
equipped with a CCD array detector: this apparatus (Fig. 1) reduces the acquisition time of a single spectrum to the fraction of a second and, due to its small size and weight (Fig. 3), it may be used as a component of a portable instrument, so as to allow also in situ measurements.
For this purpose, the optical bench may be simply replaced by the optical fibers F (Fig. 1, from the laser to the sample S') and F' (that collects the radiation emitted by S').
However, the use of the "heavy" apparatus is still essential for highly accurate measurements, which are not always required by identification problems.
FIELDS OF APPLICATION
The PL investigation of natural and synthetic materials, carried out with the apparatus described above, has been applied in many different fields, which are described below.
I)Works of art :
The identification of materials, in particular pigments, used in works of art (frescoes, sculptures, etc.) is important for art's history, and it also may help in choosing the most suitable materials for restoration [2, 4, 5].
In most cases several complementary techniques are used, the present tendency being toward non-destructive or at least micro-destructive (i.e. requiring a very small sample) methods.
Many blue pigments found in frescoes or other objects have been investigated by means of PL spectroscopy [2, 4].
Problems of interpretation can arise when the luminescent centers of a certain material are not necessarily associated to the colour, as in the case of the blue rock called "lapis lazuli" and its main mineral, lazurite.
Moreover, spectral PL assignment could be a complicate task, since a single raw material is often made of different phases, and final products could be made in turn of several raw materials, as in the case of the Maya blue (a synthetic pigment developed by the Mayas around
the VIII century A.D.) .
II) Gems :
Jewelry represents a peculiar class of handicraft [1, 6], in which gems often play a primary role. These can be made of minerals (sometimes subjected to physical or chemical treatments), or imitations, or synthetic analogues, having chemical composition and many physical properties very similar to the ones of the natural gemstones.
PL spectra of natural rubies and of most natural sapphires (varieties of corundum, namely Al2O3) exhibit in general a well resolved line system, related to Cr+3. Many imitations, widely used even for important jewels, consist of natural spinels (MgO.Al2O3), having markedly different spectral features.
As a matter of fact, although some properties (colour, refractive index) of many gems can be reproduced by synthesis, PL spectroscopy allows to distinguish not only between similar gems, but sometimes also between minerals and their synthetic analogues.
The use of PL spectroscopy, possibly through "tracing" metal ions, is specially advisable for the investigation of mounted gems or jewels of unusual shape or size .
III) Industrial products :
In many fields native impurities cause problems for their active control and determination. Nevertheless they may offer some opportunities, since they can be used as "tracers" to follow the steps (e.g. crystallization) of an industrial process.
Fig 4: PL spectra of three samples of glass containing different amounts of Fe2O3:(a) 0.034%, (b) 0.038%, (c) 0.069%.|
Electrical and optical properties of semiconductors  are determined by the whole crystal rather than by localized centers, so that such properties are extremely sensitive to impurities and strongly depend on the measurement temperature, which must be kept very low and stable. It has been demonstrated that the growth conditions (in particular temperature) under which a vapour-phase synthesis is carried out strongly affect the photoluminescent behaviour.
The purity control is particularly relevant when recycled materials are employed as raw materials. For example, fertilizing glass-ceramics  show an intense PL spectrum related to the presence of several metal ions. The most evident features were ascribed to the "uncommon" Mn(V), peculiar to the raw phosphate material also. Such information may be used in designing new materials .
PL spectroscopy is also sensitive to surface properties, hence to particle size . Therefore methods and conditions of synthesis can be selected in order to design a product having the desired properties.
On the other hand, its high sensitivity to compositional, structural and morphological properties makes PL spectroscopy not always applicable as a quantitative technique. Nevertheless, whenever samples may be produced with good geometrical and superficial homogeneity, reliable quantitative results can be obtained, for instance [3, 11] in the investigation of the iron oxidation state in industrial glass. Fig. 4 shows the PL spectra of samples containing different amounts of Fe2O3 varying from 0.034% to 0.069%.
These glasses are particularly suitable in order to check the potentialities of PL spectroscopy from a quantitative point of view, since every spectra exhibit the same shape, allowing a simplified procedure, where the measurements of the emission intensity are done at a fixed wavelength. The main band centered at about 900 nm is associated to Fe3+. For each glass, the maximum height of this emission band is correlated with the Fe3+ concentration. The trend appears quite clear, even though the effects of the concentrations of Fe2+ and Fe3+ (both present in any industrial glass) on the photoluminescence behaviour cannot be considered as fully independent of each other, in fact Fe2+ ions are able to absorb in a significant extent the Fe3+ luminescence.
SYNTHESIS AND USE OF SUITABLE REFERENCE SAMPLES
One of our current tasks consists of making the information obtained in a given case by PL spectroscopy as much transferable to different situations as possible. The full comprehension of the luminescent behaviour often requires the synthesis of suitable reference samples, which should provide particular chemical and structural features.
Fig 5: PL second order's spectra of a ruby (a) and a synthetic crisoberyl (b).|
Fig 6: schematic view of the Verneuil's apparatus: (1) mechanism for the flow of the powder, (b) flame, (c) mechanism for lowering the growing crystal.|
Under a qualitative point of view, the PL behaviour of a luminescent centre (characterized by a given atomic species, oxidation and coordination state) changes as a consequence of even slightly differences in the chemical and structural ion's environment. Fig. 5 shows, for example, that the second order's well resolved line system of Cr+3 is different in a ruby (a) and in a synthetic crisoberyl [(b), namely BeO·Al2O3], even though their host lattice are very similar.
In this case the information's transfer is relatively easy to be managed,
and the synthesis of references samples may be done, for example,
using different chemically highly pure host lattices (in the case of crystalline materials) doped with the desired concentration of the requested ion, providing some spectral lines that can be used for reference to clearly identify the presence of the same ion in different materials.
The use of PL spectroscopy as a quantitative tool is more complicated, as written above [3, 11]. In this case we managed to characterize glasses that could be used, under the same class of materials, as reference samples for quantitative PL analysis, but many problems are still unsolved when dealing with crystalline materials, due to the PL's high sensitivity to too many compositional, structural and morphological properties.
Within a comprehensive study concerning the synthesis (using different techniques) and characterization of simple and mixed oxides, we are presently developing a Verneuil apparatus, schematically shown in Fig. 6,
which will allow us to synthesize materials with all those features that permit their use as reference samples. The samples should in fact:
On the other hand, we are also currently studying the PL behaviour of some synthetic spinels grown by means of flux techniques.
- be big enough to easily permit any cut or physical treatment required for the analysis (e.g. polishing),
- contain a reduced amount of non intentional impurities, avoiding their deleterious contribution to the PL spectra,
- contain the luminescent ion in known and different concentrations, and hopefully in the oxidation and coordination state desired.
The PL spectroscopy allows a wide range of application, from the works of art to the industrial processes. The full comprehension of the luminescent behaviour often would require the use of suitable reference samples, the synthesis of which is, under certain points of view, quite complicated. The correlation of the reference samples' spectra with the ones of the materials investigated, together with the use of versatile instrumentation, that can be adapted to each particular employment (e.g. for in situ measurements) will increase the usefulness and easy application of PL spectroscopy's tool in both industrial and handicraft's fields, and will also be of great help for fundamental purposes.
- Carbonin S., Visonà D., Rizzo I., Favaro M.L., Pozza G., Ajò D. Inst. of Phys. Conf. Ser. (1998), 152, section G, 827-830.
- Ajò D., Chiari G., De Zuane F., Favaro M.L., Bertolin M. V Internat. Conf. Non-Destructive Testing, Microanal. Methods and Environm. Eval. for Study and Conservation of Works of Art (Budapest, September 24-28, 1996). Proceedings, 33-47.
- Ajò D., Polato P., Colombrino V., De Zuane F., Pozza G., Boll. Staz. Sperimentale del Vetro (2000), in print.
- Chiari G., Ajò D., Reyes-Valerio C., Virdis C., Pozza G., De Zuane F. VI Internat. Conf. Non Destructive Testing and Microanalysis for the Diagn. Conserv. Cultural and Environm. Heritage (Rome, 17-19/5/99). Proceedings, AIPnD and ICR, Rome, 1999, vol. 3, 1717-1726.
- Pozza G., Ajò D., Chiari G., De Zuane F., Favaro M.L. Sci. Techn. Cult. Herit., in print.
- Carbonin S., Ajò D., Rizzo I., De Zuane F. J. Gemm. (2000), 27, 30-31.
- Favaro M.L., Rossetto G., Ajò D., Zanella P., Torzo G., Camporese A., De Zuane F. Mater. Sci. Eng., 1994, B28, 224-227.
- Ajò D., Favaro M.L., Pozza G., Barba M.F., Callejas P., Arzabe J.O. J. Mater. Sci. ,1997, 32, 4217-4220.
- Barba M.F., Callejas P., Ajò D., Pozza G., Bettinelli M. Ceramic Engin. Sci. Proc., 1997, 18, 22-27.
- Tessari G., Bettinelli M., Speghini A., Ajò D., Pozza G., Depero L.E., Allieri B., Sangaletti L. Appl. Surface Sci. 1999, 144, 686-689.
- Ajò D., Polato P., De Zuane F. Bol. Soc. Españ. Ceram. Vidrio, 1998, 37, 315-318.