Life After People: Shaky Bridges | History

HISTORY

Jan 15, 2010

Snapping cables send these once vital links of commerce crashing into the frigid waters below. Shop Life After People today! http://www.shophistorystore.com/ HISTORY®, now reaching more than 98 million homes, is the leading destination for award-winning original series and specials that connect viewers with history in an informative, immersive, and entertaining manner across all platforms. The network’s all-original programming slate features a roster of hit series, epic miniseries, and scripted event programming. Visit us at HISTORY.com for more info.

Alliterative Abstract:

“Stop! Serpentinite!” Says Sensationalist Stanford Sage

“Bunkum!” Bellows Better Berkeley Brain

Walking in San Francisco, near Land’s End, we Geology students were shown a crumbly rock
seemingly made of mud and dust. This we were told, was Serpentinite –

The decomposed stone was shockingly easy to turn to dust in the hand. 15 years in the wind and rain will

Serpentinite – from Geology 21A Handout Weathered Serpentinite (from Sonoma)
K. Wiese S. McGuinney
“The South Tower of the Golden Gate Bridge,” we were told, “is built on Serpentinite.”Dr. Lawson, or: How I Learned to Stop Worrying and Love the Bridge – page 2
Brief Background Befitting Bridge Beginners
The Golden Gate Bridge was completed in 1937, but planning was well under way ten years prior, and
construction began in 1933. It was the longest and most complicated suspension bridge ever built at the
time, and it required new techniques and technologies which were unproven. The most public controversy concerning the feasibility of bridge regarded the placement of the foundation– called a pier – of the south
tower (to the left, below)

(Strauss, 1937)

In order to make the central span of the bridge shorter, and thus lighter, more stable, and less expensive,
Chief Engineer Joseph Strauss proposed placing the south tower 340 meters into the strait. There were
numerous unprecedented engineering challenges associated with construction in the roughly 25 meter deep
water in the Golden Gate, but the most lasting controversy concerned the quality of the bedrock into which
the piers would be built. The bedrock at the south pier is comprised of serpentinite. Serpentinite was known
then to crumble to dust when weathered. Bolstering the lay-geologists intuitive concern about this material
underlying the bridge, one of the foremost geologists of the early 20th century voiced vigorous reservations
about that bedrock.

Serpentinite is metamorphic rock primarily comprised of serpentine, which forms from olivine
metamorphosed by hot water. This tends to occur at seafloor spreading centers. Serpentine has relatively
low density and often rises to accrete at the surface. (Wiese) It can be turned by exposure to the elements
into an astoundingly weak stone which can be broken up and crushed by hand.

Serpentinite is not a very common surface rock – less than 1% of the Earth’s surface is Olivine or
Serpentinite. It does, however, serve as the bedrock along the south Golden Gate Strait. “At Fort Point, [the
serpentinite] consists of two gently east-dipping sheets at least 50 meters thick…. The serpentine sheets
consist of rounded blocks of massive bastite-bearing serpentine sheathed in intensely sheared serpentine.”
(Wahrhaftig, “39)

There are a few counter-claims to the general consensus that serpentinite is a dreadful foundation.
Serpentinite is not always weak – when kept dry, it can be a hard stone.The New York City Department of
Buildings considers Serpentinite a class 1b Medium-Hard foundation bedrock, with an allowable bearing
pressure of 40 tons per square foot– the second highest classification. (NYC-DOB, Table 1804.1)
“Serpentine, contrary to popular belief, is not especially prone to landsliding everywhere, because the
large serpentine body of Potrero Hill is almost completely free of landslides.” (Schlocker, 86)

That said, serpentinite is notorious for its poor quality as a foundation:

“Serpentine being recognized as an unsatisfactory and occasionally unacceptable foundation material. In fact, rock engineering literature warns the engineer: ‘Dams should not be placed on serpentinite” and that the material is “…unusable and unsuitable as the foundation for a dam.’”(Glawe & Upreti)

“The south pier stands on serpentinite, a notoriously weak and slippery rock – no cause for confidence.” (Alt & Hyndman, 141)

Ultimately, what becomes clear is that Serpentinite can vary in its characteristics: In an article comparing two
large projects on serpentinite, Glawe and Upreti wrote: “[Some] serpentinite can be extremely closely
fractured, with chlorite and talc infillings. The serpentinite exhibits a flaky and cataclastic rock texture… and
could be easily excavated with a pick and the remaining small blocks… could be disintegrated by hand to
sandy gravel size.” This could also describe the serpentinite atop Land’s End, as demonstrated to our
Geology 21 class.

Glawe and Upreti continue: “Why did the serpentinite of the Kusan-3 dam site exhibit higher strength than
reported elsewhere for serpentinite? ….Varying high strength values for serpentinite may result from
differences in localized lithologic factors, micro and macro structures, mineralogical compositions and
variations of interlocking of smaller grains….
If kept dry, Serpentinite can – as the continued existence of the septuagenarian Golden Gate Bridge attests –be a firm foundation stone. In 1934, this was not certain.
Bailey Willis was trained as a mechanical and civil engineer in the 1870’s. He worked as an engineer and
geologist for railroads and joined the U.S. Geological Survey in 1884. His contributions to structural geology
– regarding the formation the Appalachian Mountains in particular, led him to broad renown in geology
circles. His subsequent travels and leading work in China advanced the understanding of Chinese geology
and paleontology considerably. By the late 1890’s, he was Chief Geologist of the U.S. Geological Survey. In
addition to his scholarly works, he wrote popular books about his travels to China and Argentina. He became
head of the Stanford Department of Geology in 1915. He retired as a professor in 1922, but remained
actively involved with civil engineering and geology in California. He worked on municipal water supplies,
earthquake preparation, and, in the 1920’s, became a critic of the engineering of the Golden Gate Bridge.
(Blackwelder)

Willis conducted a years-long campaign to defeat the effort to fund and construct the Golden Gate Bridge at
its eventual location. He first wrote privately to Lawson, whom he’d known for years, and Strauss. At the
height of his campaign, Willis wrote a two-full-newspaper-page letter published in a San Francisco weekly –
The Argonaut in mid-September, 1934. Willis wrote, among other things“The southern anchorage and the south pier are founded upon a mass of sheared rock involved in a system of minor faults and consequently unstable to a degree likely to endanger the structure. The rock is serpentine and is subject to landslides, as may be seen in the immediate vicinity. Slides have occurred under natural conditions. the probability of their occurrence has been increased by blasting and would be gravely augmented by the weight of the structure it is proposed to erect on the foundation of the south pier. Such a slide would, to a greater or less extend, block the entrance to San Francisco harbor, change the tidal prism, and consequently the level of the tides, and would seriously affect the future of the city as well as cause the loss of the bridge.”
Bailey Willis, quoted in the Chief Engineer’s Report of 1937.

This public airing of Willis’s concern, along with his request that work be halted immediately and the bridge
reengineered to have a pier drilled 250 feet into the bedrock, instead of 20 feet, was unsurprisingly not
welcomed by the bridge design team. Chief Geologist Andrew Lawson charged Willis with being a professional alarmist.” (Dvorak)

Willis was very dramatic, saying the south pier was positioned “on a ‘pudding stone’ of serpentine.” He
wrote “even vibrations from the San Andreas Fault might cause the pier site and its 200,000 pound load to
slide into the channel.” He added that in a large, 1906-sized earthquake, the weight of the bridge would cause the base of the serpentine to shear, and the whole bridge would collapse into the strait. (Dvorak)

Willis was not the only credible opponent to the location of the bridge. Engineers J.B. Pope, W.J.H.
Fogelstrom, and Professor Charles Wing testified in court that “U.S. Coast and Geodetic Survey reports
proved that the rock formations under the proposed south pier could not withstand the load…” and that
Dr. Lawson, or: How I Learned to Stop Worrying and Love the Bridge – page 5
costs would be quadruple Strauss’s estimate. They were “establishment engineers… with strong local
reputations. (Starr, p66)

Bridge Geologist Lawson himself wrote in 1930 – before Willis’s efforts gained wide notoriety – that the
south pier of the bridge would “have to be designed to depend upon the dead load rather than upon the
tensile strength of the rock.”

Ultimately, Willis’s concerns were vigorously rejected by the engineering team, and his competence was
questioned – noting the 50+ years since his earning an engineering degree, and his lack of professional
experience as an engineer. His concerns were also rebutted with condescension in the Chief Engineer’s
Report of 1937, published in direct, if delayed response to Willis’s public pronouncements.

Andrew Lawson was, by the 1930’s the preeminent geologist in the country. He had authored the definitive
geological report on the 1906 earthquake, discovered and defined the extent of the San Andreas Fault, and
been instrumental in identifying the Franciscan Complex.

In the early 1930’s, when questions began to arise concerning the placement of the pier for the planned
bridge, the Chief Engineer hired Lawson to investigate and ultimate support the location of the bridge.
Lawson directed borings into the depths of the piers.

In contrast to Willis’s press-friendly protestations, Lawson was predisposed to use carefully chosen phrases
which were prone to misinterpretation by the general public. “Lawson required some coaching… to express
his findings as strongly as possible, without academic qualification.” (Starr, 94).
Lawson was noted as having described Willis’s criticisms of the Bridge locale as “a fine example of a boogyboo dragged in by the ears from the recesses of a vivid imagination to scare people. The astonishing thing
about his present attack upon the stability of the Golden Gate Bridge is that he should have restrained
himself so long.”

How the Pier Was Built

Building an enormous concrete foundation for a bridge in 60-80’ deep water with fog, shipping, tides and
storms was something that had never been done before. There were many innovations in material science,
mathematical modeling and general bridge design which were scaled up and put into practice for the Golden
Gate Bridge.

First, a temporary bridge was built from Fort Point to the pier location. A large elliptical dam was built – a
concrete bathtub the size of a football field (300’ by 155’), and extending far higher than the mean surface
level of the strait. This was necessary because the tide through the Golden Gate is disruptively forceful: up to
65000 metric tons per second, with the water moving at up to 10 km/h one way then the other. Those tides
routinely fill the water with blinding silt, and can toss boulders most vigorously. (Starr, 96)

A caisson – a building designed to be sunk to facilitate construction underwater – was built in a nearby
drydock and floated into place before the cofferdam was completed. The caisson was a very complicated
construction in its own right. It had air locks, cage elevators, airtight refuge chambers and more. This
caisson took over a year to build. Once the caisson was in place, but before the cofferdam could be
completed around it, a storm blew in, severely damaged the caisson, and threatened to damage the cofferdam
and undo a year’s work. With more winter storms coming, the engineers decided to just pump all the water
out of the cofferdam, and work without a caisson at all.

Photo Credit: E.C. Mensch, 1935

When the cofferdam was complete, the bay water was pumped out, so the silt at the bottom of the bay could
be removed. The suspect serpentinite was exposed. Deep holes were drilled into the serpentinite, and
Lawson determined that this was unusually solid, unfractured serpentinite. Still, the construction team had to
excavate twice as deep as Lawson had originally estimated… but only one fifth as deep as Willis requested, to
reach stone of sufficient quality to support the weight of the bridge. “In December 1934… Lawson
descended into an inspection well over 100 feet beneath the sea surface and reported that “the rock of the
entire area is compact, strong serpentine remarkably free from seams . . . when struck with a hammer, it rings
like steel.” The fender used to prepare the south tower foundation now serves to protect the pier from stray,
fog-bound ships.” (Dvorak)

The inspection wells (figure 5) were 15’ diameter bells at the bottom of 4’ diameter holes drilled deep into the
serpentinite. (E.C. Mensch, Ch. 7)

The concrete pier (illustrated below) was built of steel-reinforced concrete was built after the serpentinite
bedrock had been excavated 15 meters below top of the serpentinite layer.

(Illustration: Popular Science, Mar 1931, p24)
Note that this drawing, which predated
construction of the bridge by several years, does
not reflect the additional depth excavated for
additional support. The south pier extends about
fifteen meters (instead of six) below the bay floor.

SPACER

This greatest challenge met, the Golden Gate Bridge was ultimately constructed under budget, with a largely
laudable safety record. That it still stands at 77 years of age is fairly good evidence the pier is not as doomed
as Professor Willis thought.

Some thoughts on why this might be:
Willis’s initial complaints about the bridge were written in 1927, in response to an early design by Joseph
Strauss “which depended on the brute strength of rigid steel beams and thousands of tons of unyielding
structure.”. That design was subsequently abandoned in favor of a lighter, more flexible design conceived by
Leon Moisseiff (Johnson & Leon) and calculated by a team led by Charles Alton Ellis, who authored the
standard text on the mechanics and mathematics of framed structures. (Starr, 88). Ellis spent 20 months
mathematically testing the bridge design. His conclusions were that it would be sufficient to withstand the
tides, the winds, earthquakes, traffic, and its own weight. (Strauss, Appendix A) Note that Ellis is not credited
here, though he was chiefly responsible for the calculations (Starr, 100)

As compelling as the Willis-as-Cassandra narrative is, warning an oblivious public of disaster in vain – Willis’s
arguments were quite aggressively and thoroughly discredited in the Chief Engineer’s report. It should be
noted that a copy of The Argonaut with Willis’s polemic could not be acquired in time for publication. The
Golden Gate Bridge District has done a fine job preserving the counter-argument.Appendix I: 2006 Topography of the Golden Gate Strait under the South Tower of the bridge.

http://pubs.usgs.gov/sim/2006/2917/sim2917.pdf

The pier going up, almost ready for the steel towers to begin construction, 1935. (Mensch)

Bibliography – can be found on the original document.  Click Here

When Resonance Attacks

On July 5th, a 39-story office/shopping center building in Seoul, South Korea started to shake rapidly. For ten minutes vertical tremors violently rocked the building, causing an immediate evacuation of the premises. After the shaking subsided…

By Kurt Poropatich

Aug 3 2011, 7:48am

ShareTweetSnap

On July 5th, a 39-story office/shopping center building in Seoul, South Korea started to shake rapidly. For ten minutes vertical tremors violently rocked the building, causing an immediate evacuation of the premises. After the shaking subsided, engineers began a lengthy process to determine the cause of the incident. Eliminating earthquake and windstorms, the culprit they landed on was bizarre to say the least: an aerobics class of 23 people on a mid-level floor. Their Tae Bo workout was apparently twice as intense that day, making their fancy footwork synch up with the building’s structural resonance.

Resonance causing structural failure is rare, but when it happens, it tends to affect bridges. The Broughton suspension bridge in the UK and the Angers bridge in France are classic illustrations of mechanicalresonance, both of them brought on by soldiers crossing in lock-step. (The famous collapse of the Tacoma Narrows bridge has often been described as the result of resonance caused by strong wind gusts. But in fact it was brought down by the effect of aeroelastic instability, which caused the energy of the bridge in motion to feed back on itself until it collapsed.)

The Tacoma bridge is falling down

The Broughton suspension bridge in the UK and the Angers bridge in France are classic illustrations of mechanical resonance, both of them brought on by soldiers crossing in lock-step.

Millennium Bridge, 2000

More recently, London’s Millennium bridge was shut down in 2000 after a surge of pedestrians created oscillating waves across the deck of the bridge. Like pushing a child on a swing at the correct moment so that her motion is amplfiied, resonance’s effects are caused by the well-timed push of each individual wave. Push the swing at the right moment, and you could send the kid flying; amplify the vibrations in a building’s structure enough, and you can push it to failure.

spacer

Earthquakes could be triggered by sound waves that ‘fluidize’ faults

22 Sep 2015

Earthquakes may be triggered by sound waves that reduce the friction between rocks at a geological fault. That is the conclusion of an international team of researchers that has done laboratory-based experiments that mimic the process. The results support the “acoustic fluidization” theory of earthquake triggering, which seeks to explain the unexpected weakness of some faults. The research could also help to reveal how aftershocks are generated at great distances from the earthquakes that precede them.

Large earthquakes are often followed by a number of smaller tremors – called aftershocks – that can occur in geological faults thousands of kilometres from the epicentre of the original earthquake. While aftershocks are well documented, exactly how they are triggered remains a mystery.

The theory of acoustic fluidization suggests that the seismic waves generated by the initial earthquake could create sonic vibrations that help to trigger subsequent fault movements. These vibrations affect the small grains of rock at the interface between two plates at a fault. The idea is that the vibrations cause the grains to behave collectively like a fluid, lowering friction in the granular material and causing the plates to slip past each other.

Bed of grains

In their new study, the researchers used a simulated fault system to investigate whether acoustic waves may indeed be able to trigger earthquakes. Their model fault was composed of two rough, pressed-together plates, between which lay a bed of spherical grains.

When a shear stress was applied to the model system, the researchers observed the fault undergo periodic slips, as expected. Sound waves at certain frequencies were then fired at the system and this caused the fault movements to occur much sooner than when no sound was applied. The team also found that the premature slips were more likely to occur with sound waves at a characteristic range of resonant frequencies corresponding to waves bouncing back-and-forth within the fault. This observation is in line with acoustic-fluidization theory.

“Acoustic fluidization reduces the confining pressure and the system becomes abruptly unstable, promoting a transition to an unjammed fluid-like state,” explains team member Eugenio Lippiello of the Second University of Naples. “When this occurs, an earthquake nucleates.”

Spontaneous sound

Lippiello and colleagues also discovered that the same resonant sound waves emerge spontaneously from the model fault, even when no external sound is applied. This happens a short time before a fault movement. “Even in the stick phase [when the fault is motionless], the granular medium is never in a frozen state,” says Lippiello. Instead, each grain oscillates weakly around its own centre and the sound waves are generated “when oscillations of individual particles synchronize to the resonant frequency”. Accordingly, acoustic fluidization may also help to explain why the measured rate of slips on real-life faults is higher than would be anticipated, based on studies of rock-on-rock friction.

“This work moves forward the idea that the surprising weakness observed on faults during earthquakes is caused by strong vibrations close to the fault plane,” comments Jay Melosh of Purdue University, who originally proposed the acoustic-fluidization process but was not involved in this study. “These waves are generated in the core of the fault as sliding starts and – just as a car moving too fast over a rough road can skid uncontrollably – the vibrations briefly offload the normal friction and allow the fault to slide as if it were greased.” Melosh commends the team for demonstrating the effectiveness of acoustic fluidization in faults with a granular core.

Other trigger candidates

However, Emily Brodsky of the University of California, Santa Cruz, remains cautious about the results. Brodsky, who was not involved in the study, points out that there are many candidate mechanisms for triggering aftershocks. “I am not sure that the observational evidence thus far supports the assertions about acoustic fluidization being the relevant mechanism for earthquake triggering by seismic waves,” she says.

With their initial study complete, the researchers are now looking to explore in more detail the mechanism through which the grain oscillations are synchronized. At the same time, the team is also planning to explore the role of heterogeneities in granular material on the overall behaviour of the fault system.

spacer

The SAFOD drillsite is located 14 km northwest of Parkfield in central California (Fig. 1), along the creeping section of the San Andreas fault (SAF). In 2005, drilling successfully crossed the active trace of the SAF at ,3 km vertical depth12, where the measured temperature is ,112 uC (ref. 13). The drillhole terminated in sedimentary rocks of the Great Valley Sequence (K. McDougall, personal communication) east of the fault. Since then, a portion of the well
casing has been actively deforming in response to creep on a fault strand12. Serpentine was identified in X-ray diffraction patterns of cuttings14 collected at the eastern margin of the zone of active deformation (Supplementary Fig. 1). Aeromagnetic surveys15 indicate the presence of a flat-lying slab of serpentinite at .3 km depth on the northeast side of the fault (Fig. 1). This body may be $2 km thick, and it abuts the fault for 50–60 km (ref. 15). The serpentinite slab is probably part of the Coast Range ophiolite, the oceanic basement on which the sediments of the Great Valley Sequence were deposited. Serpentinized ultramafic rock has a relatively low density compared to the overlying rock column, and a fault intersecting such a rock unit provides the pathway for the migration of serpentinite to shallower depths16. The Table Mountain serpentinite17 east of Parkfield (Fig. 1) is an extrusive body that formed as a result of the diapiric rise of low-density serpentinite from the deeply buried slab along faults that served as ‘fissure feeders’17. The serpentinite associated with the active trace in the SAFOD drillhole14 and outcrops of serpentinite18,19 fault gouge (Fig. 1) suggest that the same process is occurring along the SAF.
Serpentinite has been suggested as a possible cause of creep, because of its close association with creeping faults in central and northern California6–8. The SAF creeping section coincides with the mapped extent of Coast Range 

Figure 1 | Distribution of serpentinite along the SAF creeping section. a, The creeping section lies between areas of the fault that ruptured during great earthquakes in 1857 and 1906. Serpentinite occurs in rare surface exposures of the fault18,19 and in the probable active trace of the fault encountered at ,3 km vertical depth in the SAFOD drillhole12,14. The extrusive serpentinite at Table Mountain17 is derived from the same serpentinite body that abuts the fault on the northeast side at .3 km depth15. b, Recently updated creep rates measured at distances of 10 m to 1 km from the fault1 . Total offset rates along the San Andreas system in the creeping section are considered to be between 28 and 34 mm yr21 (ref. 1).

ophiolite and overlying Great Valley Sequence rocks on the northeast side of the fault7,8,20. Ultramafic
rocks of the Coast Range ophiolite are variably serpentinized. The most extensive serpentinite body along this section is the one east of SAFOD15 that is associated with the highest creep rate (Fig. 1). On the basis of aeromagnetic and gravity surveys, a slab of serpentinite 1–1.5 km thick at $3 km depth extends northeastwards from the SAF a few kilometres southeast of San Juan Bautista to the Calaveras fault around Hollister20.

Serpentinite continues at somewhat greater depth20 east of the Calaveras fault in that area. Other, smaller masses
of serpentinite are present at $2.4 km depth between the San Andreas and Paicines faults7 . The Calaveras–Paicines faults creep at rates of 6–10 mm yr21 south of Hollister.

The creeping section provides the best evidence for a weak SAF2–5. Because the creeping section is characterized by aseismic slip and microearthquakes, the apparent weakness of this segment cannot be explained through some dynamic weakening process accompanying a major earthquake, as it can for the locked sections. The restrictions on shear strength in the creeping SAF imposed by heat-flow22,23 and stress-orientation24,25 data are delimited in Fig. 2; also included are the frictional strengths of synthetic fault-gouge materials9–11,26 prepared by grinding and sieving rock or mineral separates. The strength experiments were conducted in a triaxial machine fitted with
an internal furnace, at various combinations of temperature, confining and fluid pressure, and sliding velocity. For a given mineral to control the behaviour of the creeping section, it must be very weak as well as characterized by stable shear. The frictional properties of the serpentine minerals do not satisfy the weakness criterion and under
certain conditions do not satisfy the stability criterion. The serpentine minerals lizardite and antigorite9 are not substantially weaker than granite26 under hydrothermal conditions (Fig. 2). Chrysotile satisfies the heat-flow constraint to depths of ,3 km, but its strength increases substantially at greater depths9,10. Furthermore, all three
serpentine minerals show both velocity-weakening (strength is most abundant in the 3,325 m MD sample, which also contains nearly all of the sheared talc-bearing grains identified thus far. Talc is only a minor component of the serpentinite (#2–3%), although this soft mineral may have been preferentially lost during drilling (Fig. 3d).

Figure 2 | Shear strength versus fault depth. Shaded fields indicate the constraints on the strength of the SAF based on heat-flow22,23 and stressorientation24,25 investigations. Shear-strength data plotted for granite26, serpentine minerals9–11 and talc11 assume a temperature gradient of 30 uC km21 and a hydrostatic fluid-pressure gradient. At depths #3 km, both chrysotile and talc satisfy the heat-flow constraint, but chrysotile becomes substantially stronger at greater depths. The talc data represent a sliding velocity of 365 mm yr21 . Given the characteristic velocitystrengthening behaviour of talc11, its shear strength at #30 mm yr21 (Fig. 1) may be even lower.

We examined serpentinite grains from the washed SAFOD cuttings that were collected at ,3 m intervals during drilling. Polished grain mounts were prepared from cuttings samples for analysis with an optical microscope, scanning electron microscope (SEM) and electron microprobe. The serpentinite contents of the bulk cuttings,
estimated from point counts of thin sections, exceeds 2% by volume in the interval 3,319–3,350 m measured depth (MD), with a spike of ,8% in the 3,325 m MD sample (Supplementary Fig. 1). A powder X-ray diffraction pattern of a separate of serpentinite grains shows prominent peaks consistent with lizardite and chrysotile, the two low-temperature serpentine minerals. No relict olivine or pyroxene has been found. On the basis of the common occurrence of both mesh texture after olivine and bastite texture after pyroxene, the original ultramafic rock was probably a harzburgite16, similar to the Table Mountain serpentinite17. The pseudomorphic mesh and bastite textures have been extensively modified by recrystallization, brecciation and shearing.

The serpentinite contains numerous calcite- and some quartzfilled veins, possibly resulting from focused fluid flow within the fault zone. Talc replaces serpentine minerals along the vein walls (Fig. 3a, b), and it fills narrow cracks that extend into the serpentinite from the wider veins. Talc also forms along the foliation in sheared serpentinite grains (Fig. 3c, d). The talc-forming reaction is: 

Mg3Si2O5(OH)4 1 2SiO2 5 Mg3Si4O10(OH)2 1 H2O serpentine……………………………………………………. talc

The SiO2 comes from the dissolved silica content of heated ground water (Fig. 3a, c, d) and from quartz deposited metastably in veins (Fig. 3b). Talc is stable relative to the assemblage quartz 1 serpentine throughout the stability range of serpentine27. The veins and shears with which talc is associated overprint all other textural features in
the serpentinite grains, suggesting that the talc is of recent origin. Talc is most abundant in the 3,325 m MD 

Figure 3 | Talc occurrences in serpentinite grains. Backscattered-electron SEM images of talc-bearing serpentinite grains from cuttings collected at 3,325 m MD. a, Talc (Tc) replacing serpentine minerals (Sp) adjacent to vein calcite (Cc). b, The reaction of serpentine and vein quartz (Q) to produce talc. The reaction results in a decrease in the volume of solid phases, consistent with the concentration of pores between the talc and quartz. c, Talc forming along the foliation in a sheared serpentinite grain. d, Talc in sheared serpentinite. Talc commonly appears at the edges of grains, perhaps because the serpentinite preferentially breaks along the weaker talc during drilling. Scale bars, 50 mm.  LETTERS NATURE| Vol 448|16 August 2007 796 ©2007 NaturePublishingGroup

 sample, which also contains nearly all of the sheared talc-bearing grains identified thus far. Talc is only a minor component of the serpentinite (#2–3%), although this soft mineral may have been preferentially lost during drilling (Fig. 3d).