Chain multiplicity fault

Burgers vector and dislocation line analysis of naturally deformed aug/tc-enstatite crystals indicate activation of many slip systems such as (100)[001], 110 1<110>, 110 1<112>, (100)[010], (010)[100], (010)<101>, and 110 < 111 >, the first two being the most active [311,312]. The study demonstrates that most dislocations are dissociated and stacking faults are produced that can be interpreted based on the complex structure of these chain silicates. In naturally deformed augite from a pyroxenite with lamellar exsolution [lOl] dislocations in (010) combine single to double chains, yielding so-called chain multiplicity faults [316]. 

Mechanical (101) [101] twins have been identified in experimentally deformed hornblende single crystals, as well as dislocations on the (100)[001] slip system [333,334]. In hornblendes from naturally deformed rocks dislocations on (hkO) planes were documented, mainly [001] screws [335-338]. A systematic investigation of dynamically recrystallized hornblende from a high-temperature shear zone discovered microstructures typical of dislocation creep, with subgrain boundaries and free dislocations [313]. The primary slip system is (100)[001] consistent with experimental results. Secondary, slip systems are (010)[100] and 110)5<110>. There is evidence for cross-slip of [0 01] screws producing heUcal microstructures [Fig. 13(b)]. Amphibole structures are intermediate between pyroxenes and sheet silicates and indeed chain multiplicity faults have been described [339] and transitional structures may be facilitated by movement of partial dislocations [340]. 

The most common accident models today explain accidents in terms of multiple events sequenced as a forward chain over time. The events included almost always involve some type of failure event or human error, or they are energy related (for example, an explosion). The chains may be branching (as in fault trees) or there may be multiple chains synchronized by time or common events. Lots of notations have been developed to represent the events in a graphical form, but the underlying model is the same. Figure 2.4 shows an example for the rupture of a pressurized tank. 

---