The Ring Monstrance from the Loreto treasury in Prague: handheld Raman spectrometer for identification of gemstones

Abstract

A miniature lightweight portable Raman spectrometer and a palm-sized device allow for fast and unambiguous detection of common gemstones mounted in complex jewels. Here, complex religious artefacts and the Ring Monstrance from the Loreto treasury (Prague, Czech Republic; eighteenth century) were investigated. These discriminations are based on the very good correspondence of the wavenumbers of the strongest Raman bands of the minerals. Very short laser illumination times and efficient collection of scattered light were sufficient to obtain strong diagnostic Raman signals. The following minerals were documented: quartz and its varieties, beryl varieties (emerald), corundum varieties (sapphire), garnets (almandine, grossular), diamond as well as aragonite in pearls. Miniature Raman spectrometers can be recommended for common gemmological work as well as for mineralogical investigations of jewels and cultural heritage objects whenever the antiquities cannot be transported to a laboratory.

This article is part of the themed issue ‘Raman spectroscopy in art and archaeology’.

1. Introduction

In the last few decades, Raman spectroscopy has been confirmed as an excellent tool for mineralogical studies of all kinds [1–5]. Nowadays, Raman spectra are frequently recorded and interpreted at the same time that new minerals are described and accepted by the International Mineralogical Association [6–8]. By this method, further uses of this technique for discrimination and identification purposes are facilitated and welcomed by mineralogists as well as non-specialists. Elsewhere, Raman spectroscopy is also widely used for non-destructive unambiguous identification of minerals and pigments in the areas of the arts and cultural heritage [9–12]. This is well documented by the increasing number of reports of successful applications within these fields.

Interestingly, during the last decade increasingly smaller, lightweight instruments have appeared, and are increasingly used in such diverse areas as forensic science, geoscience, art and archaeology. Because they are very light and portable, battery-operated Raman instruments have recently been applied to discriminate minerals from several groups [13–15].

Some successful practical applications of miniature Raman spectrometers were also described under outdoor outcrop conditions [16] or at archaeological sites [17,18].

Raman spectroscopy is now routinely used as a primary or screening method in gemmology, and for studying gemstones mounted in historical objects under laboratory conditions (golden chalices from the Einsiedeln Abbey, Switzerland, crafted in the seventeenth century [19] and the Reliquary Cross and Dorothy Monstrance from Basle Cathedral, Switzerland, of the late Gothic period [20]) or even in situ (Heinrich's Cross from Basle Cathedral dating from the late Middle Ages [21], the sceptre of the Faculty of Science of Charles University, Prague, Czech Republic [22] and the silver Torah shield [23], Prague) without the necessity to move or remove the artefacts. Recently, Raman microspectroscopy was used for characterization of emeralds [24] and rubies and their imitations [25].

Critical issues on the use of portable instruments have also been highlighted: possible effects of sunlight during short measurements [26], light absorption effects and fluorescence on dark minerals [27] and systematic displacements of measured Raman shifts (mainly at low wavenumbers) [28]. Vandenabeele et al. [29] addressed the possible effects of the distribution of minor phases in powders on Raman spectroscopic semi-quantification of traces.

Here, for the first time, portable Raman tools were used for the rapid mineralogical screening of gemstones mounted in religious jewels held within the framework of the Loreto treasury and collection in Prague. This was allowed during the time frame of recent restoration work and reorganization of the Loreto treasury exhibits in preparation for an exhibition entitled ‘The Hidden Face of the Loreto Treasure’.

The Prague Loreto is a remarkable complex of religious buildings mainly from the seventeenth and eighteenth centuries—a place of pilgrimage with a captivating history. Inside, there is hidden a spectacular treasury, with a collection of jewels and religious objects assembled through the centuries by the Order of Minor Chapucines.

The expansive decorative frontal façade of the Loreto, with a clock tower from which the carillon tunes may be heard on the hour, shall clearly not go unnoted, and is well known worldwide. But it is only when one walks through the frontal building, entering into the intimate patio, that one finds oneself next to the ‘Loreto’ itself, i.e. the Loreto Chapel proper, the so-called Santa Casa. It is a copy of the ‘Holy House’ of Nazareth worshiped in the Italian Loreto, as the place in which the Miracle of Incarnation had occurred (here, archangel Gabriel announced to the Virgin Mary that she would conceive and become the mother of Jesus, the son of God, marking his Incarnation). Construction of the Prague Loreto Santa Casa began on 3 June 1626, at the instigation of Baroness Benigna Katharina von Lobkowitz. Lavishly decorated with reliefs on the outside after the Italian prototypes, it is the spiritual centre of the entire place of pilgrimage. Standing freely in the courtyard area, it is surrounded by a multi-storey cloister with corner chapels. The pilgrimage Church of the Nativity of Our Lord, one of the most impressive surviving Rococo interiors in Prague, is another part of the complex occurring in the axis of the courtyard, beyond the Loreto Chapel. The Loreto and the associated Loreto treasury is a phenomenon known not only to native-born Prague inhabitants but also to a majority of those visiting Prague for sightseeing. The Loreto treasury arose gradually over the course of the seventeenth and eighteenth centuries; the interiors were partly renovated in the nineteenth century. In the 1950s and 1960s, a new treasury was built, which is accessible to the public. For a long time, the Loreto treasury was perceived through its best and most brilliant pieces, such as the diamond and ring ostensoria and other works made of precious metals and stones, which are mentioned in the majority of guides and other publications [30,31].

The Loreto treasury room is where a considerable part of the famous Loreto treasure is exhibited. Next to the St Vitus Cathedral treasure, it is the most valuable ecclesiastical treasure in the Czech Republic. The most valuable piece, in terms of both the materials and artistry, is the famous Diamond Monstrance, made between 1696 and 1699 in the Viennese workshops of J. B. Khünischbauer and M. Stegner. The 6222 diamonds decorating this treasure were from the bequest of Countess Ludmila Eva Frances Kolowrat. This treasure is rarely used in the liturgy today, and then only for very exceptional occasions. The last time that pilgrims were able to venerate the Body of Christ, exposed in the gloriole of shining diamonds, was during a celebration in the autumn of 1999, 400 years after the arrival of the first Capuchins in Bohemia. No less interesting are other precious pieces: the Ring Monstrance (investigated in this report), a small household altar featuring the representation of the scene of the Nativity of Our Lord, made of ebony wood and silver; or the late medieval chalice dating from 1510.

From the set of items examined in the collection of the Loreto treasure, the so-called Ring Monstrance is the most visually attractive. Additionally, from the viewpoint of being able to add to historical information about past jewellery techniques, it is also the most significant item. Monstrances (from the Latin monstrare = to show) appeared during the thirteenth century as a category of religious objects of reverence and worship. Its creation was necessitated by the need to safeguard the casing for the sanctified (consecrated) wafer exhibited on the altar table, representing the body of Christ. Its general shape is derived from one type of medieval reliquary, a glass casing with peripheral decoration, and it is placed on a high stand. A specific technical design was then derived from the contemporary visual style. We can frequently find this architectural design, depicting elements of the vaulting system during the Gothic period, while the so-called Sun-type, with peripheral decoration of a central shell, consists of distinctive radial beams and shaped reliefs. During the Baroque period, additional decoration, often with precious stones, was quite popular.

The Ring Monstrance can be included in the list of Sun-type monstrances, as its basic shape is no different from other monstrances of this period and type. However, what makes it extraordinary when compared with the others is the manner of its decoration. This monstrance was made in 1728 in the workshop of the well-known Lesser Town silversmith, Master František Michal Rädelmayer to order for Loreto's sacristan. František Michal Rädelmayer was born in 1689 in Freysinghof in Bavaria, and died in Prague in 1748. He was employed as a silversmith's journeyman in Prague from 1717, and was accepted as a guild master in 1724. He primarily worked for ecclesiastical institutions, from which a relatively large number of his liturgical objects are still preserved (see [32]). The monstrance body has no master or any official mark apart from some additional tax marks from the beginning of the nineteenth century, and thus the identification of the manufacturer and its dating was only possible on the basis of the inscription on the edge of the base. ‘G[e]macht worden per Franz Michael Redelmeir in Prag Anno 1728’. The inscription is done by stippling, and its readability is thus difficult; in the older literature, the monstrance is dated to the year 1748. Advanced dating has been made possible by the recent discovery of archival records.

This paper describes the application of a miniature Raman spectrometer to confirm the heretofore presumed presence of gemstones, glass and pearls in the seventeenth century monstrance. This is a small, very light-weight, palm-sized Raman device, used for the first time to investigate a complex historical artefact with 65 mounted jewels and 324 stones.

2. Material and methods

(a). Objects

(i). Minor artefacts

The studied palla (P-090) is square-shaped, richly embroidered with gold thread and decorated with many red stones. Along the perimeter, there were 14 small enamel images with saints, and in the middle was an enamel heart with a picture of the Virgin Mary with Infant Jesus. The jewellery parts were affixed onto a textile base. Interestingly, the large garnets used were probably originally worn strung on a chain. The lunula (P-174) is composed of one large clear stone, and is richly decorated with tiny gemstones. It is a fragment of a complex piece of jewellery, probably a hairpin (P189), and is again decorated with a variety of clear stones and a central red silicate.

(ii). The Ring Monstrance

The monstrance is made of silver of an unknown fineness; individual parts are chased and hammered out, some minor parts are cast, and individual construction components are connected together by soldering, rivets and screwed joints. The entire area of the surface is gilded by amalgam. A vase-shaped stem, with an outlined notching decorated with cast figures of angels, is mounted on an oval and slightly raised base. It carries its own decorative casing for the consecrated wafer. The circumference of the glass casing for the wafer is decorated with a chased relief, with a figure of God the Father under a canopy placed over the glazing; the outer circumference of the casing consists of a set of radial beams, of unequal lengths, forming an almond-shaped design. The total height of the monstrance is 915 mm.

Hitherto the monstrance had been considered to be very well made; however, it is actually a standard specimen, not differing from its historical environment. However, what fundamentally distinguishes it is its decoration with precious stones, enamelled elements and carved cameos, many of which had been used in earlier jewellery and later incorporated into significant portions of the design. For the most part, there are the bezels of rings, which is how the monstrance acquired its name. A considerable amount of jewellery was collected in the monastery's treasury store during earlier periods, mostly donations dedicated as votive gifts by pious believers. It was not possible to use these for the liturgy in any meaningful way, so the usual practice was to dismantle the jewellery and use the resulting metals and precious stones for other purposes. In this case, this did not happen, although some jewellery was adapted for new usages: in the case of the rings, by cutting out their hoops; in the case of the fine enamelled decorations, by their division into smaller parts. Nevertheless, it is still possible to discern the essence of the original jewellery. It should be noted that, in this period, damaged and unsatisfactory contemporary fashion jewellery was redesigned without any hesitation; however, material evidence of this kind from the sixteenth and seventeenth centuries is preserved in limited numbers. Nevertheless, in this case, the jewellery was retained and represents a very interesting insight into cultural history, as well as making a valuable contribution to our understanding of historic jewellery techniques and the art of gem cutting.

(b). Instruments

Two commercially available handheld and one palm-sized Raman spectrometer were used in this study, specifically an Inspector Raman by Delta Nu, a FirstGuard by Rigaku and a FirstDefender RM by Thermo Scientific.

The handheld Inspector Raman instrument weighs 1.9 kg. It is equipped with a 785 nm diode laser, with a maximum output power of 120 mW, and a thermoelectrically cooled CCD detector. It provides Raman data over the 200–2000 cm⁻¹ wavenumber range, with a spectral resolution of 8 cm⁻¹. This instrument is remotely controlled via a USB cable-connected laptop. It allows the operator to select from a wide range of settings for measurements, including: more detailed power output selection, the number of accumulations and the duration of accumulation times. The spectra were typically recorded in situ at the following setting: 1–5 s accumulation time, and usually just one accumulation for rapid identification. The laser power was set to ‘high’ for most measurements; in some cases, the laser power was set lower to avoid fluorescence and detector saturation. The estimated output laser power is close to 120 mW for the ‘high’ setting, and close to 60 mW for the ‘medium’. This instrument was used for analyses of both the treasury artefacts and the monstrance.

The FirstGuard is equipped with a 532 nm laser, but is also a handheld pistol-like instrument. It weighs 2.7 kg, and is equipped with a thermoelectrically cooled CCD detector. The maximum laser power output at the source is 60 mW. It operates in the spectral range of 200–3000 cm⁻¹, with a spectral resolution of 10–15 cm⁻¹. It includes a resistive touch display for control, and allows a wide range of experimental settings. In this study, the spectra were typically collected at the following setting: 1–10 s accumulation time, usually one accumulation, and laser power at 1–50 mW. This instrument was used for analyses of the treasury artefacts.

The FirstDefender RM instrument weighs only 0.8 kg and is representative of the palm-sized instrument category. It is also equipped with a 785 nm diode laser and has a maximum output power of 250 mW and a thermoelectrically cooled CCD detector. It provides Raman data over the wavenumber range of about 200–2500 cm⁻¹, with a spectral resolution of 7–10.5 cm⁻¹. This instrument is operated using buttons, and has a built-in LCD screen. In this work, the instrument was operated in an automated program, where the parameters of Raman spectrum acquisition are calculated for each sample individually, without the operator's input. The parameters are nevertheless saved in the data files; for example, the acquisition times varied from tens of milliseconds to a maximum exposure time of 10 s. This instrument was used for analyses of the monstrance.

All measurements were carried out by shielding the specimen and the instrument's contact area using a black cloth. The Raman spectra were exported into Galactic *.SPC format. The spectra were compared using GRAMS AI v. 9.1 (Thermo Electron Corp., Waltham, MA, USA). The Raman spectra were not subjected to any data manipulation or processing techniques; generally, they are reported as collected, with the exception of the spectra taken with the FirstDefender RM instrument, where cutting out of excitation laser line artefacts was necessary. A baseline correction was applied where mentioned.

The measurements presented, and the mineralogical results obtained, are just some examples of what was achieved during a short inspection survey. In making these measurements, it was possible to evaluate a wide range of minerals in a variety of artefacts. Further stages of the survey will focus on the possibility of assessing the origins of some of the gems that have not yet been studied. Miniature Raman spectrometers allow for very fast and unambiguous identification of gemstones in situ. It is obvious that such tools might very well be used in further explorations of other historic artefacts decorated with or made from minerals, gemstones and other natural products of different origins.

3. Results and discussion

(a). Loreto treasury artefacts

Selected jewellery pieces as well as cultural artefacts such as crowns that are part of the Loreto treasury collection were studied by handheld Raman spectrometers to confirm/disprove the presumed types of gemstones. The list of the selected analysed pieces is given in table 1, and detailed photographs can be seen in figure 1. It was discovered that the variety of gemstones and other materials, confirmed by Raman spectroscopy, was rather limited in the samples studied. The most common stones were red-coloured garnets of almandine composition, and a relatively high number of diamonds were confirmed. These were readily distinguished from the colourless stones such as crystal quartz and colourless silica glass that were also detected. The last material used for adorning the jewels was the pearls, with the aragonite Raman signal confirmed. Diagnostic Raman bands for the minerals are reported in table 2, together with the reference values.

Table 1.

List of studied artefacts from Loreto treasury.

artefact	approximate age	origin	description of materials
palla P-090	before 1766	central Europe	cardboard, silk cloth/textile, bobbin lace-made trimming of golden thread, enamel, garnets, glass imitation of gems, watercolour images on parchment
cross pendant N-041	first half of the eighteenth century	central Europe	silver, clear glass (imitation of crystals)
lunula P-174	between 1728 and 1755	Prague (?)	gold, smoky quartz, almandine, diamonds
hairpin fragment P-189	eighteenth century	central Europe	gold, silver, garnet, diamonds

Figure 1.

Selected pieces of the Loreto treasury: (a) hairpin fragment P-189, (b) cross pendant N-041, (c) lunula P-174 and (d) palla P-090.

Table 2.

Raman band positions of identified materials on the Loreto treasury artefacts.

artefact	colour	instrument	Raman band positions (cm−1)	Raman band positions reference (cm−1)	identification
palla P-090	dark red	Delta Nu	321, 344, 369, 498, 556, 631, 864, 917, 1041	323, 342, 370, 475, 500, 556, 630, 863, 916, 1038 [33]	garnet (almandine)
	red	Delta Nu	∼600 (broad), ∼1360		glass
cross pendant N-041	colourless	Delta Nu	∼430, ∼770, 980sh, 1034, ∼1200, ∼1360		glass
		Rigaku	∼400, ∼780, 980sh, 1040, 1130
lunula P-174	dark red	Delta Nu	345, 352, 475, 498, 554, 867, 916, 1041	323, 342, 370, 475, 500, 556, 630, 863, 916, 1038 [33]	garnet (almandine)
		Rigaku	348, 555, 917
	colourless	Delta Nu	1334		diamond
		Rigaku	1332
	smoky yellow	Delta Nu	208, 354, 393, 464, 806	206, 264, 354, 390, 450, 464, 697, 796, 808, 1069, 1162 [34]	smoky quartz
		Rigaku	211, 356, 407, 466, 809, 1081, 1157
hairpin fragment P-189	colourless	Delta Nu	1331		diamond
		Rigaku	1332
	dark red	Delta Nu	364, 557, 641, 863, 918, 1039	323, 342, 370, 475, 500, 556, 630, 863, 916, 1038 [33]	garnet (almandine)
		Rigaku	555, 860, 920

Number in bold or in italics denotes a strong or a weak Raman band, respectively. sh denotes a shoulder.

Garnet was the most common gemstone type that was identified on the Loreto treasury artefacts. All of these appeared to have a dark red/brown colour, some of them were cut and other pieces were in the form of large garnet beads, presumably originally worn on a string as a necklace. Garnets of close to almandine composition were identified on the following artefacts: a hairpin fragment P-189 (figure 1a), lunula P-174 (figure 1c) and palla P-090 (figure 1d). Garnets belong to the nesosilicates group, with the isolated SiO₄ tetrahedra in the form of orthosilicate anions connected to interstitial metal cations. The strongest Raman bands (signature bands) in the Raman spectrum of garnets are located at wavenumbers around 918, 555 and 345 cm⁻¹, which can be attributed to the (Si–O) stretching, (Si–O) bending and libration modes, respectively, in our garnet samples; these are in good agreement with the values given by Kolesov & Geiger [33] for pure almandine (916, 556 and 342 cm⁻¹). Translation modes due to either or X²⁺ (metal) give rise to Raman bands typically close to or below 200 cm⁻¹ and, therefore, could not be observed due to the spectral range limitations of the handheld instruments. The garnet spectra are discussed below. They are generally very similar, with the notable exception that the changes in intensity ratios of bands located at around 555, 365 and 345 cm⁻¹ can be explained by the orientation of the crystals; therefore, the intensity of the total symmetric modes A_1g is most affected. Additionally, some spectra contained broad features, probably luminescence related, while others had a perfectly flat baseline.

Another type of precious stone that was quite widely present in the studied samples were diamonds. Some of these were quite large, but, according to stringent present-day criteria, their colour, clarity and cut are not exceptional. Diamonds were identified on the following artefacts: hairpin fragment P-189 (figure 1a) and lunula P-174 (figure 1c). Diamond is an allotropic high-pressure modification of graphite with the highest form of symmetry. The sp³ carbon atoms form the diamond lattice, and, according to the selection rules, only one mode is Raman active and is manifested by a narrow band at 1332 cm⁻¹. The Raman signal of diamond was typically the strongest and was the most easily collected when using the handheld instruments. We were able to confirm 13 gemstones as diamonds using both of the instruments by the only Raman band in our spectra located between 1328 and 1334 cm⁻¹; these are reported in figure 2. The ease of detection of the diamond Raman signal was exploited with great advantage for the discrimination of the great number of gemstones of the monstrance; this is covered in the next section.

Figure 2.

General view of the monstrance.

One of the important tasks was to discriminate between colourless and coloured silica glass pieces that were used to imitate gemstones or as replacements for the missing pieces. In the samples from the Loreto treasury that were analysed, the most common were colourless glass, probably used for diamond imitation, with red-coloured glasses sometimes being used alongside the garnets. Light blue-coloured glasses were also distinguished. Red-coloured glass, used as a garnet imitation, was identified on the palla P-090 artefact (figure 1d). Colourless glass, used as a diamond imitation, was used as the sole adornment material on the cross pendant N-041 (figure 1b). Raman spectra of the glasses are quite different from the other crystalline gemstones. Broad features are prominent in the Raman spectra due to both the Raman signal of the amorphous glass melt as well as the luminescence phenomena, for example from impurities or pigments that were manifested when the 785 nm laser was used. Nevertheless, two different sets can be differentiated in our samples (figure 3): the Raman spectra of colourless glass; and those of red-coloured glass that are simpler, with the signal arising from the SiO bonds overlapped by luminescence features quite possibly from the red pigment. A relatively pure Raman signal of glass was obtained using the 532 nm excitation when analysing the colourless pieces of glass. The following wavenumber values of bands in the Si–O stretching region were observed: the strongest band at 1040 cm⁻¹ (Q₂ mode), 980 cm⁻¹ and 980 cm⁻¹ shoulder (Q₁); other bands due to the Si–O stretching components are located at 780 (Q₀) and 1130 cm⁻¹ (Q₃); and an additional broad band at around 300–500 cm⁻¹ can be clearly observed, which is attributed to the Si–O bending vibrations (figure 3). Band assignments were adopted from Colomban et al. [35].

Figure 3.

Raman spectra of gemstones recorded on artefacts, as seen in figure 1.

Several types of SiO₂, such as varieties of pure quartz (crystal, amethyst, smoky quartz, red quartz), quartz and chalcedony materials coloured by impurities (jasper) or other microcrystalline and cryptocrystalline (carnelian) forms of SiO₂ in gemmological qualities, were used for adorning the pieces of the Loreto treasury. They are all easily distinguished from other types of gemstones by the characteristic Raman band of quartz located at 464 cm⁻¹. The second strongest band located at around 206 cm⁻¹ can only partially be seen due to the limited spectral range of the handheld instruments. A smoky quartz was identified on the lunula P-174 (figure 1c) in the studied artefacts.

Using the 785 nm instrument, pearls were confirmed as an aragonite polymorph of CaCO₃ by the characteristic Raman bands located at 1086 cm⁻¹ and 703 cm⁻¹, which are attributed to the ν₁(CO₃) symmetrical stretching vibration and ν₄(CO₃) symmetrical bending vibrations, respectively.

(b). The Ring Monstrance

The significance, priceless historical value as well as simply the large dimensions, elaborate design and the vast amount of gemstones that adorn this religious artefact (figure 2) merit a separate section in this work. As both the manipulation of and the time that this artefact could be studied outside of the safe room were quite restricted, the analytical part of this study was planned with a goal to analyse as many gemstones as possible in the shortest possible time for each individual analysis. This was generally achieved, because, for the unambiguous identification of gemstones, the experimental settings of the handheld instruments including short accumulation times (under 5 s) and low numbers of accumulations (mostly single accumulation) provided relatively good quality spectra with the Delta Nu instrument. The Thermo Scientific palm instrument was used in the automatic mode, where the duration and number of accumulations were determined by the instrument, based on the quality of the Raman signal arising from the gemstones. The time for one analysis was typically quite short (a few seconds) for diamonds; other stones (such as beryls) took several tens of seconds to analyse (the accumulation times were in the range between 25 ms and 10 s). In this manner, over 200 Raman spectra were acquired in under 5 h using two handheld Raman instruments.

In table 3, typical examples of each of the identified gemstones, along with their corresponding detected Raman bands and reference values from the literature, are given. Images of selected individual parts mounted on the monstrance are shown in figure 4; corresponding spectra are reported in figure 5, with no spectral manipulation, including the fluorescent background. The significant or diagnostic Raman band wavenumber values are also given for each mineral. Spectroscopic identification of gemstones from five interesting regions on the monstrance will be discussed in the following text, and both detailed photographs and corresponding Raman spectra are given. The gemstones are numbered using the exact same numbers that were assigned to them by the curators when they performed the detailed photodocumentation and categorization of the monstrance. These numbers are used in table 3 and in the associated photographs (figure 4a–e) and Raman spectra (figures 6–10).

Figure 7.

Spectra of ruby, grossular, diamond and ‘red quartz’ identified in figure 5b (dashed lines are expected peak positions of the diagnostic bands).

Figure 8.

Spectra of diamond and amethyst identified in figure 5c (dashed lines are expected peak positions of the diagnostic bands).

Table 3.

Raman band positions of identified materials on the monstrance.

gem placement in adornments	colour	instrument	Raman band positions (cm−1)	Raman band positions reference (cm−1)	identification
‘bird and rays’ 2,27 (figure 4a)	colourless	Delta Nu	1335	1332 [29]	diamond
‘floral’ 6 (figure 4b) ‘portrait’ 1,26 (figure 4c) ‘crossed heart’ 6 (figure 4d) various 7,13,15 (figure 4e)		Thermo	1332
‘bird and rays’ 1,3 (figure 4a)	purple-pink	Delta Nu	380, 417, 595, 643, 749, 1040, 1206	378, 417, 430, 451, 578, 645, 750 [36]	corundum (ruby)
‘floral’ 1,5 (figure 4b) various 1,14 (figure 4e)		Thermo	379, 419, 536, 576, 643, 750
‘crossed heart’ 2 (figure 4d)	dark blue	Delta Nu	420, 578, 750	381, 418, 644, 574, 644, 749 [37]	corundum (sapphire)
‘floral’ 18 (figure 4b)	red	Delta Nu	252, 281, 377, 551, 628, 828, 882, 1010	247, 280, 373, 376, 550, 630, 827, 880, 1007 [33]	garnet (grossular)
		Thermo	246, 278, 374, 549, 826, 881, 1005
‘star’ 4,13,14	red	Delta Nu	348, 369, 498, 555, 596, 634, 865, 917, 1041	323, 342, 370, 475, 500, 556, 630, 863, 916, 1038 [33]	garnet (almandine)
		Thermo	354, 370sh, 505, 559, 640, 861, 917, 1042
‘portrait’ 16 (figure 4c) various 10 (figure 4e)	light violet	Delta Nu	211, 266, 400, 464, 796, 1065	206, 264, 354, 390, 450, 464, 697, 796, 808, 1069, 1162 [34]	quartz (amethyst)
		Thermo	208, 263, 355, 402, 465, 807
‘crossed heart’ 1 (figure 4d)	green	Delta Nu	240, 324, 396, 680, 1067	325, 400, 526, 686, 1068 [38]	beryl (emerald)
		Thermo	330, 385, 400, 536, 682, 1066, 1196, 1385
‘floral’ 19 (figure 4b)	pearly	Thermo	206, 704, 1084	209, 702, 706, 1084 [39]	pearl (aragonite)

Number in bold or in italics denotes a strong or a weak Raman band, respectively. sh denotes a shoulder.

Figure 4.

(a–e) Details of five different types of adornments containing various types of gemstones.

Figure 5.

Examples of Raman spectra of gemstones that were identified on the Ring Monstrance, and are displayed in detail in figure 4 (spectrum taken using the Delta Nu (DN) or Thermo Scientific (TM) instrument).

Figure 6.

Spectra of diamond and ruby identified in the bird and the halo on figure 4a (dashed lines are expected peak positions of the diagnostic bands).

Figure 10.

Spectra of amethyst, ruby and diamond identified in figure 5e (dashed lines are expected peak positions of the diagnostic bands).

The initial idea in this study was to either confirm or disprove that the ‘diamond-like’ gemstones are actually genuine diamonds, because they were used for adornments of this monstrance in great numbers. From over 80 randomly selected ‘diamond-like’ gemstones, all were confirmed as real by their Raman spectra, and thus it can be safely concluded that all the presumed diamond stones are genuine and not substitutes. The diamonds were confirmed in the ‘bird and rays’ adornment, the large stone (2) between the bird's wings and a randomly selected smaller stone (27) from the multitude below the bird and on the halo (figure 4a). In the ‘floral’ adornment to the top-right side of the monstrance (figure 4b), the triangle-shaped stone (6) was confirmed as a diamond, and the eyes of the frog are also diamonds. In the ‘portrait’ adornment (figure 4c), the two groups of colourless stones on the right side, in the rings, are diamonds (stones 1 and 26 confirmed). In the adornment depicting a ‘heart crossed with arrows, crown and wings’ (figure 4d) the rectangular stone (6) was confirmed as a diamond. In the several different adornments seen in figure 4e, diamonds decorate the cameos (7), the large stone (13) and also the cross with the centre stone (15).

The second most common type of gemstone that was confirmed on the monstrance was ruby. However, the colour was purple-pink and so, according to some classifications, they could be determined as pink sapphires. The corundum crystal Al₂O₃ has seven Raman active modes: the A_1g modes give rise to bands located at 417 and 645 cm⁻¹, and the E_g modes give rise to bands located at 378, 430, 451, 578 and 750 cm⁻¹ [36]. However, the intensity of the bands is quite dependent on the crystal's orientation of the corundum (ruby) gemstones towards the laser beam. Thus, in our samples, only two signature Raman bands of the corundum located at 418 and 750 cm⁻¹ were detected in the majority of the corundum spectra. Sometimes the band located at 646 cm⁻¹ had a much higher intensity than the 750 cm⁻¹ band. Additional weaker Raman bands located at 379 and 576 cm⁻¹ were detected by both instruments in several spectra. In general, there was great variety in the general quality of the Raman spectra of rubies recorded by both instruments. Spectra of some rubies contained substantial amounts of fluorescence/luminescence, displayed as both a steep baseline and broad bands. In the ‘bird and rays’ adornment (figure 4a), the coloured stones (1 and 3) are confirmed as rubies; in the ‘floral’ adornment (figure 4b) the purple-pink stones (1 and 5) in the centres of the flowers are confirmed as rubies. In the ‘heart crossed with arrows, crown and wings’ adornment, which has the richest variety of gemstones (figure 4d), the purple-pink stones are rubies, and the dark blue stone (2) is a sapphire. This is the only sapphire that was identified and spectroscopically confirmed on the monstrance. The quality of the spectra of the sapphire taken by the Delta Nu instrument were generally only mediocre (see figure 9) even in the repeated analyses. However, at least two diagnostic bands at 420 and 750 cm⁻¹ are clearly visible and are in a good agreement with the reference values [37]. In the group of adornments displayed in figure 4e, stones 1 and 14 are confirmed as rubies.

Figure 9.

Spectra of emerald, diamond and sapphire identified in figure 5d (dashed lines are expected peak positions of the diagnostic bands).

Amethyst, the purple-coloured variety of crystalline quartz, was identified by the typical quartz spectrum, with the most intense band located at 464 cm⁻¹. This spectrum had an almost flat baseline. Amethysts were identified as the 14 purple stones encircling the ‘portrait’ adornment (figure 4c), with stone 16 confirmed. Amethyst is also confirmed as the solitary stone (10) in the adornments displayed in figure 4e. However, at least two other quartz stones of red colour were identified, and the spectra of these contain huge amounts of luminescence features, probably due to colouring agents; the most prominent are features at around 1370 and 1860 cm⁻¹. This is interesting because a similar 1360 cm⁻¹ broad feature was present in the red-coloured glasses in some of the artefacts studied in the previous section; but, in this case, with this type of stone on the monstrance, the quartz band at 464 cm⁻¹ is still clearly visible. One such stone (11) with colour similar to the grossular (18) was confirmed in the ‘floral’ adornment just below the triangle-shaped diamond (figure 4b).

Garnets were only used on the monstrance as a few single relatively large stones, with an orange-red colour. Using the Raman spectroscopy, we discerned two types of garnets, grossular and almandine, based on the most intense Raman band located at around either 880 or 917 cm⁻¹, and due to the (Si–O) stretching vibration in the A_1g mode. Handheld instruments typically provide a good to very good quality of spectra when analysing garnets. Additionally, the Raman spectra of grossular-type garnets contained several features in the range 1100–1800 cm⁻¹ that can be attributed to the luminescence of some trace elements, such as rare earth metals, rather than the Raman signal. The spectra of almandine-type garnets are entirely without these features. The characteristic of the measurements (i.e. quick identification, where a sufficiently high quality of spectra is seldom achieved) does not permit chemometrics, for example. Stone 18, below the ‘floral’ adornment (figure 4b), was confirmed as a grossular garnet.

One green gemstone, emerald, was confirmed by both handheld instruments. The Raman spectrum of emerald (beryl) is characterized by two signature bands located at 1068 and 680 cm⁻¹, and these are attributed to the Si–O symmetric stretching vibration and the Be–O symmetric ring deformation vibration, respectively [38]. Weak intensity Raman bands at 324 and 396 cm⁻¹ (attributed to the ring vibration) were detected in the spectrum taken by the Delta Nu instrument. An emerald, stone 1, was confirmed in the ‘heart crossed with arrows, crown and wings’ (figure 4d).

Pearls were confirmed as pieces located to the left of the ‘floral’ adornment (stone 19 in figure 4b) using the palm Thermo instrument based on the Raman bands located at 1084, 703 and 206 cm⁻¹.

4. Conclusion

Miniature lightweight portable Raman spectrometers and palm-sized devices allow for fast and unambiguous detection of common gemstones mounted in complex jewels. This was documented studying several complex religious artefacts and the Ring Monstrance—lustrous antiquities from the Loreto treasury (eighteenth century). These discriminations are based on the very good correspondence of the wavenumbers of the strongest Raman bands. However, frequently, portable instruments do not permit the systematic collection of all of the faint Raman signals. In general, however, the very short laser illumination times and efficient collection of scattered light were sufficient to obtain strong diagnostic Raman signals, as recorded on polished homogeneous diamonds, oxides (corundum), quartz and silicates (rubies, emeralds and sapphire). In order to obtain acceptable Raman spectra, even on diminutive parts of the jewels containing gemstones, simple hand positioning was acceptable without using a tripod. Miniature Raman spectrometers can be recommended for common gemmological work as well as for mineralogical investigations of jewels and cultural heritage objects whenever the antiquities cannot be transported to a laboratory.

Authors' contributions

J.J. conceived the study, designed the study, coordinated the study, participated in data collection and analysis and drafted the manuscript. A.C. carried out the molecular spectroscopic work and participated in data analysis. M.B. assisted in logistics, assisted in data collection and helped draft the manuscript. P.B. assisted in data collection, assisted in logistics and helped draft the manuscript. J.K. assisted in data collection and helped draft the manuscript. All authors gave final approval for publication.

Competing interests

The authors declare that there are no competing interests.

Funding

Partly funded by the institutional support MSM0021620855 from the Ministry of Education of the Czech Republic (to J.J.).