MICRO-ARCHITECTURE

Head for your local shop-ping mall and you may well fi nd yourself looking at vast areas of polished rocks (FIG. 1). Some of them are probably old friends used architectur-ally in many countries. On the left of my picture, we see a wall clad in black lar-vikite from the Oslo Rift, Norway, with gleaming iridescent alkali feldspars. Most of the fl oor is a pale

pink granite of (for me) unknown provenance. The inlaid slabs include a brown rapakivi granite, with enigmatic ovoid feldspars, from the mighty Wiborg intrusion, southern Finland. The other rock is metamorphic, a high-grade gneiss, birthplace unknown.

The main mineral in the three plutonic igneous rocks is alkali feldspar, ~80% in the larvikite, 50% or so in the rapakivi granite, and more than 30% in the pink granite. For architectural use, it is the colour and transpar-ency of the alkali feldspar that gives each rock its distinctive appearance. From the point of view of petrology, it is easy to dismiss these differences as trivial. Most petrologists would be happy to consider all three rocks to be ‘fresh’, then analyse them for all manner of elements and isotopes, and make deductions about their sources and evolution. But the differences in appearance are not trivial. Because larvikites with iridescent feldspars are rare and pink and white granites very common, they tell us something profound about a process that has affected large volumes of the crust.

I don’t usually bring work by myself and collaborators into these little articles, but emboldened by the fact that I’ve recently taken a part in two very enjoyable workshops on fl uid–mineral reactions, organized by the guest editors of this issue and held in sunny parts of Europe, I’ll break with tradition. The differences in feldspar colour you can see in your shop-ping mall are not caused primarily by chemistry but by the underlying structure of the feldspars, features at the micrometre and submicrometre scale. These are usually described as ‘microtextures’. Perhaps ‘micro-archi-tecture’ might be appropriate in the present context. The microtextures tell us about the cooling history of the crystals, and most fundamentally, about the reactions they have undergone with fl uids.

FIGURE 2 is a transmis-sion electron microscope (TEM) image of an irides-cent alkali feldspar from a larvikite from Greenland. When it grew from magma at about 850 °C, during the Mesoproterozoic, this feldspar would have been structurally featureless, but as it cooled, over about 104 years, it went through a process of solid-state exso-lution and twinning. The stripy parts are Na feldspar, albite, and the stripes are Albite twins. The zig-zag lamellae are K-feldspar,

microcline, and each zig and each zag is a twin. It is Bragg diffraction of light by these microtextures that causes the iridescence.

Exsolution in alkali feldspar is driven by the large difference in ionic radius between Na+ and K+, and diffusion of these ions in the larvikite has occurred through an Si–Al–O framework that has remained continuous. The crystal structure has organized itself into this complex architecture to minimize elastic strains between the Na- and K-rich parts. At the atomic

scale, it is a 3-D array of individually rigid tetrahedra behaving collectively like tiny springs – springs that have remained under tension or compres-sion for 1,166,000,000 years, providing a perfect, undisturbed time cap-sule. They tell us that the crystal has escaped any interaction with fl uids over that immense span of time. It has retained almost all its radiogenic 40Ar, and I can imagine no process by which it could have changed its major element composition, or the stable isotope composition of its major elements, since the Mesoproterozoic, without releasing the little springs.

The overwhelming majority of feldspars are not so lucky. Often, at a tem-perature well below the solidus, they react with a fl uid, usually after they have passed through a stage of solid-state exsolution. FIGURE 3 shows the profound changes that occur when a cooling iridescent feldspar from a larvikite meets a fl uid, just below 500 °C. The regular ‘braid’ microtextures have been replaced by a mosaic of subgrains. As seen in hand specimen or thin section, the crystals have become turbid because micrometre-scale cavities (micropores) have formed at subgrain junctions (FIG. 4). The iri-descence has been lost and the feldspar has turned white. In thin sec-tion and in sizeable cleavage fragments, the iridescent larvikite feldspars are perfectly transparent, those from the rapakivi granite predominantly transparent, and those from most other granites variably turbid in thin section and opaque even in small fragments. Micropores are often fl uid fi lled. Johnson and Rossman (2004) calculated that the amount of water stored in fl uid inclusions in feldspars in the upper crust is about the same as that contained in all the hydrous minerals put together. The subtle pink colour of the fl oor of my shopping mall is caused by tiny hematite particles within some of the pores.

Pore formation accompanies a dissolution–reprecipitation process that occurs without change in the outline of the original crystal, which becomes a partial or perfect pseudomorph. In my larvikite example, the process is driven by minimization of strain energy without change in crystal bulk composition, a process called ‘mutual replacement’. The subgrain mosaic has lower free energy than the elastically strained fi ner-scale intergrowth. In alkali feldspars from granites, the microtextures are different, because the parent feldspars contain relatively more K. Recrystallization sometimes occurs by mutual replacement, but it can also be a true replacement reac-tion, involving (largely) Na and K exchange between feldspar and fl uid. The bulk composition of the feldspar is then not constant. Both types of replacement can be inferred for many feldspars, and the introduction of replacive feldspar is often guided by pre-existing exsolution microtextures.

Given that the feldspar in most granites and felsic gneisses is at least partly turbid, pink, grey or white, it is clear that huge volumes of the upper crust have been subject to reactions with circulating aqueous fl uids. They have recrystallized, partly or completely, probably during the later stages of their initial cooling history, but perhaps repeatedly. Common, garden-quality alkali feldspar crystals do not necessarily have bulk compositions acquired at the magmatic stage. The next time you go to the shops, look in awe at the twinkling larvikite. It is a real rarity, a truly ‘fresh’, evolved plutonic rock, an exotic part of an important story.

Ian ParsonsUniversity of Edinburgh

Johnson EA, Rossman GR (2004) A survey of hydrous species and concentrations in igneous feldspars. American Mineralogist 89: 586-600

FIGURE 1 Evidence of fl uid–rock interaction, available near you

FIGURE 2 Alkali feldspar from a larvikite with ‘braid’ microtexture. The width of this TEM image is 0.8 µm.

FIGURE 3 Backscattered electron image of braid texture, at

the left, recrystallized to a patchy mosaic of albite (black) and microcline (grey) subgrains. Black dots are micropores, causing turbidity. Width = 200 µm

FIGURE 4 TEM image of the interior of an albite patch. It is built of albite

subgrains, with micropores (white) at their junctions. Rows of white dots and light-grey rims to the pores are caused by beam damage. Width = 2.3 µm

ELEMENTS JUNE 2013238

PARTING SHOTS

ADVERTISERS IN THIS ISSUE

AHF Analysentechnik 240

Applied Spectra 240

Bruker 240

Bruker Nano 176, 240

Cambridge University Press 232

CAMECA 239

CrystalMaker 174

Excalibur Mineral Corporation 166

FEI 202

Geochemical Journal 237

Geochemist’s Workbench Back cover

Geological Society of London 210

IWA Publishing 234

JEOL 240

McCrone Microscope and Accessories 167

Rigaku Inside front cover

Savillex Inside back cover

SPECTRO 210

TSI 175

JOB POSTINGS

NanoGeoScience – University of Denmark 237

See www.elementsmagazine.org/jobpostings

Essential for ultra trace analysis in geochemistry

PFA LABWARE

www.ahf.de :: info@ahf.deOUR EXPERIENCE … YOUR PROFIT!

Tandem LA - LIBS

Visit us at Goldschmidt 2013 in Florence, Italy

Tandem LA LIBSNew possibilities

in geochemical discovery

Simultaneously measure major and trace elements along with isotope ratios!

www.bruker.com/elements

Structural and Compositional Analysis at the Microscale

ELEMENTS JUNE 2013240

from the leader in fl uoropolymer manufacturing

Savillex Corporation has been proudly serving the geochemistry community for over 35 years. Our fl uoropolymer labware products are used throughout the

world for sample digestion, separations, storage and many other applications. We understand the unique needs of geochemists and now our new line up of ICP-OES and ICP-MS sample introduction products combine the highest performance and the lowest metal background with the ruggedness and reproducibility required for routine analysis.

All of our ICP sample introduction products are designed, molded and tested in house, using only the highest purity grade PFA resins.

For technical notes, videos and to fi nd your local Savillex distributor, visit www.savillex.com.

Savillex Corporation Phone: 952.935.4100 | Fax: 952.936.2292 Email: info@Savillex.com | www.savillex.com

Nebulizers – C-Flow

• C-Flow microconcentric PFA nebulizer range with the narrowest uptake rate specifi cation

• New C-Flow 35 with uptake rate of 35uL/min +/-7uL/min

• C-Flow range for desolvators – standard fi tment on the CETAC Aridus II

• C-Flow 700d with removable uptake line for high solids applica-tions – up to 25% TDS

• All C-Flows can be backfl ushed without tools

Inert Kits

• PFA kits with Scott type chamber • True double pass design gives

lower RSDs• O-ring free end cap• Platinum or sapphire injectors• Available for Agilent

7500/7700/8800 and Thermo Element 2/Neptune

• Lowest metal background

• Highest seal integrity

“Geochemistry would not be where it is

today if it were not for Savillex Corporation.”

– Dr. Mike Cheatham, Syracuse University