Failure Analysis of the Arecibo Observatory 305-Meter Telescope Collapse (2024)
Chapter: Appendix C: Arecibo Telescope Cable Failure Mechanisms Considered by the Committee
C
Arecibo Telescope Cable Failure Mechanisms Considered by the Committee
Several material degradation phenomena produce a time-dependent loss in the cable load-carrying capability that can eventually trigger the failure of load-bearing cables that lack redundancy. Table 6.3-1 of the NASA Engineering and Safety Center report1 summarizes these modes of time-dependent material degradation. The significance of these time-dependent degradation modes is that applied stresses well below the yield strength can activate or trigger certain phenomena leading to possible failure of materials in service even though the breaking strength or even a fraction of the yield strength was never exceeded in service by operative stresses from all sources. The relevance to the Arecibo Telescope is that factors of safety expressed by the minimum breaking strength of a cable in pristine condition divided by an applied stress, assuming the original load-bearing strength of the cable connection remains the same, do not apply to aged structures and cable connections with degraded load carrying capability. This is because the net section stress may rise as the connection material degrades by uniform thinning, creep, or crack propagation, which reduces the load-bearing capacity of the connection.
Moreover, the applied stress intensity factor at growing cracks may increase at the same or fixed global stress (or the same tension on wire cables) until the fracture toughness of the material is reached. This will depend on the size and growth of stress-concentrating flaws. Pull out of zinc discussed previously and referred to herein and by NASA, Thornton Tomasetti, Inc. (TT), etc., as “slip or flow” include time dependent creep and instantaneous plastic deformation. Based on the pristine original cable strength, the applied force or tension needed to cause creep is still only a fraction of the minimum breaking strength or yield strength achieved during one-time loading to fracture. In a creep failure process, fast loading eventually occurs as a sample cross-section of load bearing area decreases, and the remaining cross-sectional load-bearing area is the last material to separate by fast fracture. This fast fracture surface can exhibit a ductile mode of fracture, such as indicated by micro-scale micro void coalescence or macro-scale shear failure.
CREEP
The creep strain rate depends on the homologous temperature, TH, given by absolute operating temperature relative to melting temperature (T(K)/Tm(K)), where K indicates Kelvin. Typical relationships have the following
___________________
1 G.J. Harrigan, A. Valinia, N. Trepal, P. Babuska, and V. Goyal, 2021, Arecibo Observatory Auxiliary M4N Socket Termination Failure Investigation, NASA/TM−20210017934, NESC-RP-20-01585, NASA Engineering and Safety Center, NASA Langley Research Center, June, https://ntrs.nasa.gov/api/citations/20210017934/downloads/20210017934%20FINAL.pdf (hereafter “NESC Report”).
form where n and m take on integer values or can be zero depending on the operative creep mechanism. The ratio of applied shear stress relative to the shear modulus is raised to the n power, giving a strong dependency on stress.
In this equation, Ai is a material constant, Di is the diffusion rate for self-diffusion of zinc in this case, ε is strain, t is time, σ is shear stress, G is shear-modulus, k is Boltzmann’s constant, T is temperature, b is the burgers vector of a dislocation, Ω is the atomic volume of the metal (zinc) undergoing creep, and d is the (zinc) grain size. In this expression, n = 2–6 in the case of power law creep (PLC) while m is often 0. There is, in this case, limited dependence on grain size. However, when Nabarro-Herring (N-H) or Coble creep prevails, n = 0 and m = 2–3. Hence, when creep occurs by these mechanisms, there is no dependence on stress as creep is driven by vacancy diffusion.
Thus, it is important to understand the prevailing creep conditions during the operation of the Arecibo Telescope. A material has different creep “regimes” based on the values of T/Tm and σ/G. as indicated on creep deformation maps.2 Each material has its deformation map. In each regime, there is a dominant mode of creep such as Coble creep, Nabarro-Herring (N-H) creep, dislocation creep (or twinning), PLC, or dislocation glide instead of creep,3 and the equation above is modified accordingly. These creep mechanisms are significant to the Arecibo Telescope failure as the creep regime dictates what factors matter and which do not. For instance, Coble and N-H creep do not depend on stress. N-H creep within sub-grains and power law creep depend not on grain size but are very sensitive to stress. NASA, TT, and the authors of this report agree that PLC was very likely operative in the zinc spelter socket at the Arecibo Telescope.4,5 Given PLC, doubling the stress increases the creep strain rate substantially when n = 4 or 6 versus n = 2. More will be said about this below.
Despite all this information, the lifetime of components in service or laboratory coupons cannot be estimated from a deformation map. Textbooks note that low temperatures near 0.5 of Tm and moderate operating stresses, coupled with long service lifetime, represent the conditions where creep should be considered in engineering design. Unfortunately, in the Arecibo Telescope design, computing a static factor of safety based on static loads/stresses and materials properties with time-dependence degradation in their load-carrying capability was unfortunate. Load carry capability then would decrease over time in service.
It should be noted that deformation maps also report deformation regimes prevalent at low temperatures and high stresses where creep may not be operative and are short-circuited by dislocation glide. These convey the deformation regimes encountered near room temperature and at high load rates, such as during a tensile test. Consider a rising load tensile test over a short period. Here, deformation is elastic at low stress, and as stress is raised, dislocation glide occurs once the yield strength is exceeded. Time-independent plastic deformation occurs with possible work hardening until higher stresses are applied, which reach the theoretical breaking strength of the material. Creep is not operative in this regime. Fast laboratory loading at room temperature does not assess creep susceptibility or behavior during a creep process. Unfortunately, the laboratory tensile testing on replicated versions of failed parts falls in this category.
Furthermore, ramifications in a socket when creep of zinc enables load transfer amongst broomed steel wires is a complex consequence of creep not factored in when using routine safety factors. The spelter socket is normally not a weak point in the suspension system, but this is true only when under fast loading at room temperature, such as in a proof test. In the deformation map regime representative of a tensile test, elastic loading, the yield and breaking load dominated by time-independent elastic and plastic deformation of the material, is encountered. The following section turns attention back to the conditions and circumstances at Arecibo.
___________________
2 H.J. Frost and M.F. Ashby, 1982, Deformation-Mechanism Maps: The Plasticity and Creep of Metals and Ceramics, Pergamon Press.
3 T.H. Courtney, 1990, Mechanical Behavior of Materials, McGraw-Hill.
4 TT Final Report.
5 NESC Report.
The cast metal anchor concept for attachment of high-tensile wire cables relies on transferring wire tension forces to the zinc and the inner wall of the cast steel socket.6 Annular, longitudinal, and radial compressive stresses are developed.7 The adhesion and friction forces between the wire and cast zinc must be greater than the friction forces between the zinc and socket wall. In this arrangement, it is well-understood that initial zinc “flow” will occur during initial proof loading as zinc is engaged and seated. This is seen in proof testing, where wire failure must occur in the strand, not the socket. Bridge hanger design documents report that the socket should be designed to develop the ultimate strength without suffering measurable creep of the zinc under load. This stipulation seems to have been disregarded in the Arecibo Telescope design and the subsequent inspections by all the structural engineers.
The existing literature consistently observes that long-term cable socket slip under long-term loading is not just a matter of zinc engagement/repositioning within the socket as during initial loading. Detailed metallurgical training and analysis are not necessary to grasp these principles. For instance, Rehm, in the journal Wire in 1977,8 wrote a subsection within the paper titled “Improving Slip and Creep Characteristics” where the second part of the subsection is labeled “the long-term slip which takes place after such repositioning if the stresses in the cone are still higher than the creep limit of the casting material.” The paper goes on to explain possible methods for reducing slip. Vinet and Roberg9 studied creep in creep-resistant Zn-Al alloys and showed the strong temperature, stress, and time inter-relationship enabling engineers to predict displacements on the order of 1 inch in 20 years at 30°C at 30 percent of the rated tensile strength for the poured metal cone-shaped socket.
In a component such as the spelter socket, permanent deformation of zinc changes the positions of wires during wire/core slippage and brings about zinc flow. Zinc flow can change the complex state of stress on the wires, such as the transfer of load to outer wires in the broom. The load transfer to the outer wires in the broom is exacerbated by non-uniformity in the broom. This is in complete agreement with NASA,10 Wiss, Janney, Elstner Associates, Inc. (WJE),11 and TT reports.12 However, what is missing in these Arecibo Telescope socket failure mechanism investigations is a cogent explanation for the irregular, non-linear timeline of Arecibo Telescope’s socket slip or core pullout. The mechanisms leading to Arecibo Telescope socket failure appear to have been arrested from 2004 to 2011, and perhaps even since initial seating in 1997. There is no evidence of increased socket pullout up to 2018, which appears to have then accelerated after Hurricane Maria, as well as after the M4N-T August 2020 cable failure and the M4-4 failure of November 2020.
The rapid increase in cable breaks is indirect evidence that supports the notion that slip was accelerated or the consequences were accelerated. The more noticeable consequences readily observed were the increase in wire breaks and the cracking in the back of socket M4-4T, as shown in Appendix E, Figure 29 of the TT Final Report.13 It is possible that time-independent plastic deformation of an episodic nature occurred post-Maria, where instantaneous plastic flow was stimulated by an increase in tension. This plastic flow would have occurred abruptly on top of continual time-independent creep. The committee feels that both processes contributed to the overall failure and the accelerated pace of wire breaks just before the December 1, 2020, failure and the Arecibo Telescope collapse. The TT Final Report did include PLC of zinc, although its implementation in the finite element analysis of socket broomed wires is unclear.14
___________________
6 I. Ridge and R. Hobbs, 2012, “The Behaviour of Cast Rope Sockets at Elevated Temperatures,” Journal of Structural Fire Engineering 3(2):155–168. https://doi.org/10.1260/2040-2317.3.2.155.
7 G. Rehm, M. Patzak, and U. Nurnberger, 1977, “Cast Metal Anchors for High-Tensile Wire Tendons,” Wire 26(5):173–180.
8 Rehm et al., 1977, “Cast Metal Anchors for High-Tensile Wire Tendons.”
9 R. Vinet and R. Roberge, 1978, “Evaluation of the Creep Resistance of Wire Rope End Attachments,” Journal of Engineering Materials and Technology 100(2):214–216, https://doi.org/10.1115/1.3443475.
10 NESC Report.
11 Wiss, Janney, Elstner Associates, 2021, Auxiliary Main Cable Socket Failure Investigation, WJE No. 2020.5191, June 21 (hereafter “WJE Report”), p. 7.
12 Thornton Tomassetti, Inc. (TT), 2022, Arecibo Telescope Collapse: Forensic Investigation, NN20209, prepared by J. Abruzzo, L. Cao, and P.E. Pierre Ghisbain, July 25, https://www.thorntontomasetti.com/sites/default/files/2022-08/TT-Arecibo-Forensic-Investigation-Report.pdf (hereafter “TT Final Report”).
13 TT Final Report.
14 TT Final Report.
CORROSION
Corrosion of zinc, steel wire, and steel spelter socket casting is a thermodynamically spontaneous process that occurred naturally on Arecibo Telescope materials listed herein in the natural Arecibo environment. The issue is whether the corrosion rates had any effect on structural integrity. Material corrosion can occur in several modes, and atmospheric attack was likely the primary mode at Arecibo. Atmospheric corrosion of steel and zinc mainly occurs by a relatively uniform corrosion process, which sometimes may include shallow pitting. Corrosivity at Arecibo is controlled by environmental factors, the alloy chemistry, and any corrosion mitigation strategies applied. Corrosivity categories can be classified based on the environment.15 Corrosion rates can then be estimated. Wet and dry cycling, time of wetness, relative humidity relative to salt deliquescence, salt deposition rates, dew point phenomena, and other atmospheric species, especially sulfur, affect corrosivity. Salt aerosols are often carried inland from seawater sources, and their presence depends on the sea state helping to create the aerosol, the distance of the structure in question from the ocean, wind conditions, time of wetness, as well as periodic rinsing by rainwater. NaCl and MgCl2, as well as sulfates, are to be expected. SO2 present in industrial applications is a potent factor.
Several publications define steel corrosion rates accurately as a function of atmospheric conditions using power law formulas expressing corrosion rate as a function of these factors.16 The results for plain carbon steel would be modified by the wire steel composition.17 Corrosion rates are modified by galvanizing.18 Corrosion penetration results in a reduced wire cross-sectional area that progressively serves as the “remaining” load-bearing area after long exposure without mitigation. Hence, a fixed applied tension load and corrosive conditions would result in greater applied stress on the reduced cable net section over exposure time, gradually increasing toward a minimum breaking stress and, finally, failure. The safety factor is thereby eventually reduced to unity should this corrosion mode continue unabated. Failure by tensile overload could be achieved by corrosion at fixed tension should the load-bearing cross-section be reduced sufficiently such that minimum breaking strength is approached. However, this process is mitigated by galvanizing and application of a zinc-rich primer to the wires.
The scenario of reduced wire cross-section affecting structural integrity was found to be unlikely at Arecibo. The spelter socket casting was too thick for any appreciable change in the load-bearing cross-section of the socket itself. Moreover, wires were well protected by zinc galvanizing, as well as a zinc-rich primer, and the sockets had humidity controls. The zinc-rich primer was periodically reapplied. The maintenance activities at the Arecibo Telescope were effective in this regard. The zinc functions as a barrier and a sacrificial anode to protect any steel wires exposed.19 Little zinc corrosion was observed in failed spelter socket cross-sections. Hence, any scenario placing uniform corrosion as the source of failure is extremely unlikely at Arecibo due to insufficient corrosion penetration and minimal loss of load-bearing areas on each subcomponent.20
The zinc-rich primer and galvanizing (ASTM Standard 586; class A, 1 oz/ft2 depending on wire diameter21 with zinc subject to specification B 6) was fairly intact at the Arecibo Telescope, meaning that this sacrificial cathodic protection was still operative at the time of failure and likely protected the steel from the aforementioned hypothetical loss in cross-section. One ounce/ft2 can protect to the point that no rusting is observed for more than 10 years in a marine environment and 20 years in a rural environment.22 The exact protection in the tropical environment of Arecibo is uncertain. Failed wires were reported to have from 0.85–1.35 ounces/ft2 of galvanized zinc.23 White corrosion products traced to zinc carbonates were observed, indicating the zinc was corroding, while
___________________
15 International Organization for Standards (ISO), 2012, “ISO 9223: Corrosion of Metals and Alloys—Corrosivity of Atmospheres—Classification, Determination and Estimation,” https://www.iso.org/obp/ui/#iso:std:iso:9223:ed-2:v1:en.
16 W. Hou and C. Lang, 2004, “Atmospheric Corrosion Prediction of Steels,” CORROSION 60(3):313–322.
17 ANSI, 2020, “ASTM G101: Standard Guide for Estimating the Atmospheric Corrosion Resistance of Low-Alloy Steels.”
18 J.W. Spence. F.H. Haynie, F.W. Lipfert, S.D. Cramer, and L.G. McDonald, 1992, “Atmospheric Corrosion Model for Galvanized Steel Structures,” CORROSION 48(12):1009–1019, https://doi.org/10.5006/1.3315903.
19 D.A. Jones, 1995, Principles and Prevention of Corrosion, 2nd Edition, Pearson.
20 TT Final Report.
21 American Society for Testing and Materials (ASTM), 1998, “ASTM A586-98: Standard Specification for Zinc-Coated Parallel and Helical Steel Wire Structural Strand and Zinc-Coated Wire for Spun-In-Place Structural Strand,” https://doi.org/10.1520/A0586-98.
22 A.J. Stavros, 1987, “Corrosion,” AMS Metals Handbook, Ninth Edition, Volume 13, ASM International, p. 432.
23 TT Final Report.
very little red rusting was seen, which likely indicates steel corrosion.24 Zinc corrosion products are indicative of a functional cathodic protection strategy. The observed presence of actively corroding zinc suggests ongoing functional cathodic protection of the steel. Under cathodic polarization, the steel corrosion rate could be reduced by a factor of up to 100. The presence of zinc galvanizing and zinc primer mitigated corrosion of the steel to the extent that reduction in a load-bearing cross-section was unlikely. However, zinc presents a “trade-off” and leads to enhanced hydrogen production from the reduction of water, which can affect hydrogen-assisted cracking susceptibility. This is discussed below.
STRESS CORROSION CRACKING AND HYDROGEN-ASSISTED CRACKING
Steel wire utilized as tension elements in cables of suspension bridges and at the Arecibo Telescope brings together all the factors necessary to render them susceptible to hydrogen embrittlement (HE), more specifically, hydrogen environment-assisted cracking (HEAC) and stress corrosion cracking. Stress corrosion cracking (SCC) is the cracking of a material produced by the combined action of corrosion and sustained tensile stress.25 To further illustrate the insidious nature of SCC, it should be noted that cracking can initiate and grow at stresses well below the yield strength, rendering a safety factor based on stress not useful. If a crack-like defect exists in a steel cable, which places the applied stress intensity factor (SIF) below the dry air fracture toughness but above the threshold SIF for SCC initiation and crack growth, failure may occur by fast fracture after a period of slow SCC growth, which reduces the load-bearing cross-sectional area. SCC susceptibility requires a susceptible material governed by composition/microstructure, applied tensile stress (flaw-free) or SIF (flaw present), and a corrosive environment. SCC and HEAC are thus material-specific. Steel susceptibility is a function of alloy composition, microstructure, and mechanical properties.
High-strength steels with high material hardness are more susceptible than lower-strength alloys.26 The cable wires in this application may be regarded as high strength, as indicated by their hardness. The stress may come from multiple sources and may be residual self-equilibrated stresses contained in the part or gravity-load induced. Removal of the tensile stress, the detrimental metallurgical condition, or the corrosive environment eliminates the conditions where SCC is operative. None of these is possible at the Arecibo Telescope.
SCC also describes hydrogen-assisted cracking, where hydrogen is produced due to corrosion. HE, when exposed to a corrosive environment, is also called SCC. In the NASA report, HE is termed HEAC or hydrogen-assisted cracking. The ASTM definition of HE states that it is cracking in the presence of hydrogen. In this application, hydrogen is produced electrochemically from water and proton reduction during corrosion of the steel wire and/or by cathodic protection when steel is protected by zinc. Steel wires at 200 ksi strength levels, such as cold-drawn UNS G10800 steel, are susceptible to HE at electrode potentials experienced when galvanically coupled to zinc, which is documented by Enos and the references within.27 A portion of the hydrogen generated is absorbed into the steel, which lowers the breaking load, tensile strain, and threshold stress intensity in case of pre-existing flaws that concentrate stress. Under static tensile loads, cracks can initiate and grow at stresses well below the yield strength. The breaking strength and ductility of steels are reduced by the hydrogen concentration developed in the alloy. HEAC depends on the hydrogen concentration absorbed and is often marked by a critical hydrogen concentration where the fracture mode changes from ductile to brittle. The breaking strength and ductility of steels are reduced by the local hydrogen concentration developed in the alloy. HE, and the broader category of SCC, is plastic strain-rate dependent. Fast strain rates would deny the chance for brittle fracture even if hydrogen is pre-charged, while slow strain rates or static loading in high-strength alloys allow it to operate. This is because hydrogen must be repartitioned by solid-state diffusion (which is slow) to the high tri-axial stress at the crack tip. This goes hand-in-hand with observing ductile fracture in wires experiencing overload and fast fracture during rapid loading.
___________________
24 TT Final Report.
25 ASTM, 1993, Annual Book of ASTM Standards: Wear and Erosion; Metal Corrosion, Vol. 03.02, ASTM International.
26 Stavros, 1987, “Corrosion,” p. 432.
27 D.G. Enos and J.R. Scully, 2002, “A Critical-Strain Criterion for Hydrogen Embrittlement of Cold-Drawn, Ultrafine Pearlitic Steel,” Metallurgical and Materials Transactions A 33:1151–1166. https://doi.org/10.1007/s11661-002-0217-z.
A fraction of the wires in the spelter socket outside the core were noted to fracture in a brittle fashion by the slow process described above. However, fast fracture at a high strain rate in the laboratory during a fast tensile test is not a viable methodology to detect SCC or HE susceptibility. The particular morphology of fracture surfaces observed at high magnification are often fingerprints or indicators of cracking mode. Fatigue, SCC, and hydrogen-assisted cracking have their own telltale flat fracture morphologies, which may be microscopically seen as cleavage or intergranular. Brittle failure of prestressing wires can also occur by longitudinal splitting.28
Arecibo Telescope steel wire of 220 ksi tensile strength subject to ASTM A 586 was utilized.29 The steel met the strength ductility of drawn AISI 1080 or UNSG10800 steel. This steel is near the eutectoid composition. Hence, the microstructures developed ultra-fine eutectoid pearlite. One hundred and sixty-eight wires were assembled using a helical arrangement. At the Arecibo Telescope, steel wires underwent fracture in the socket—but outside the wire broom in one failure and outside the zinc casting but near the socket in another case—during the final collapse.
Broken wires have been known and reported occasionally.30 Subcritical breaks occurred in many of the Tower 4 wires outside the socket core but at random positions with respect to the socket core. Whether inside or outside, steel wires were exposed to zinc and periodic moisture. Concerning HE, zinc facilitates water reduction and hydrogen entry, and the hydrogen centration in the steel is greater than what can be introduced by atmospheric corrosion alone without zinc. This is due to cathodic polarization brought about by the zinc galvanizing and zinc-rich primer in a galvanic couple with the steel. The zinc potentially increased the hydrogen content of the wires at the Arecibo Telescope. Thus, the HEAC should have been possible, but little evidence was observed. Several wires underwent slow crack growth before the final fast fracture of the remaining wires. This is indicated from cross-section examinations and was indicated by black/brown corrosion products, which suggest the cracks found predate the actual failure event. Moreover, subcritical cracks away from the primary fracture were detected upon inspection of a few wires that failed during the collapse of Tower 4.
HE should likely be removed from consideration as a primary root cause of the failure. This is because brittle cracking was indicated by fractography on only a small fraction of the wires in the outer region of the broomed wires and anywhere else. Subsequently, Lehigh University confirmed in Fritz laboratory tests that the breaking strength of steel cables and sockets during fast fracture was not degraded during laboratory tensile tests of failed sockets/cables. The Phoenix et al. paper indicates the same.31 However, it should be noted that this test was conducted at moderate to fast loading rates, which would not have detected the presence of HEAC due to its slow strain rate dependence. When it is stated that the steel wires were not weakened during service, that is a correct statement in the case of a tensile test applied during fast loading at room temperature, such as in a proof test. Fractography inspections indicate mainly ductile cup/cone failure and shear wire breaks versus HEAC. This is the expected result in the case of a final fast fracture in an overload situation at the time of the collapse. This supports the ductile wire separation argument and points elsewhere to find a primary root cause. Other arguments, analyses, and evidence are all in favor of zinc slip pull-out and zinc rupture. Hydrogen content was not measured in the steel, nor were slow strain tests conducted on harvested wires when galvanically coupled to zinc. Therefore, the overall extent of HE damage to wires is currently unknown, but the lack of fractographic evidence does not support HE as the root cause of failure.
FATIGUE
Steel wires and cables are susceptible to fatigue. Fatigue leads to failure at applied static tensile stress well below the yield point in uniaxial tension or under mixed loading modes. A small cyclic tensile stress amplitude well below the yield point but sufficiently high (i.e., above the endurance limit) triggers crack initiation and propagation in a ferrous material given enough cycles. Fatigue has three stages. These are initiation, propagation, and growth until fast fracture occurs and is brought about by a reduced remaining load-bearing cross-section due
___________________
28 Ibid.
29 TT Final Report.
30 L. Phoenix, H.H. Johnson, and W. McGuire, 1986, “Condition of Steel Cable After Period of Service,” Journal of Structural Engineering 112(6):1264, https://doi.org/10.1061/(ASCE)0733-9445(1986)112:6(1263).
31 Phoenix et al., 1986, “Condition of Steel Cable After Period of Service.”
to fatigue crack growth. Fatigue depends on the material, and a specific material’s behavior is often depicted in a plot showing maximum applied cyclic stress causing initiation, propagation, and growth as a function of the number of cycles of the load. Greater cyclic load amplitudes reduce the number of cycles to failure. In steels, the endurance limit is defined as the applied cyclic stress below which there are an infinite number of cycles needed to attain initiation and propagation. Given enough cycles, fatigue failure will occur at a lower global stress below the minimum break strength observed in a single loading to failure. Fatigue can be affected by corrosion either by changing initiation, propagation, and/or growth. Corrosion followed by fatigue test in dry air allows sampling of the effect of corrosion on fatigue, which produces surface damage and increases stress concentration. True “corrosion fatigue” is concurrent corrosion and fatigue, which is usually transgranular and indicated by markings (striations) related to crack advance as a function of number of cycles.
The potential sources of fatigue loading are numerous at the Arecibo Telescope and include vibrations, wind loading, and even day/night cycling, given applied tie-down tension and thermal expansion and contractions. Oscillations were partially mitigated with the use of dampening devices. Fatigue loading cycles tend to have a small cyclic stress amplitude relative to static stress. This level of cyclic stress has been regarded to be insufficient to trigger fatigue damage in this application. This was also concluded previously.32 The location of the fracture inside the spelter socket also argues strongly against a fatigue failure mode. At the Arecibo Telescope, fracture surfaces did not indicate fatigue as far as the usual accompanying observation of striations or markings on fracture surfaces. Fatigue striations were not observed at the Arecibo Telescope.
LONG-TERM ELECTROPLASTICITY
Troitskii and Likhtman33 introduced modern electroplasticity. Zuev et al.34,35 discussed the mobility of dislocations in zinc single crystals under the action of electric current pulses. Stashenko et al.36 reported the significant effect of current pulses on the creep of zinc single crystals during short-term laboratory tests involving high current densities. More recent papers by Conrad,37 Guan et al.,38 Lahiri et al.,39 and Rudolf et al.40 address the use of electroplasticity in manufacturing (also known as electrically assisted manufacturing).
Stashenko et al.41 reported that the “flow of conductivity electrons gives part of its energy to defects which participate in plastic deformation of the metal.” The authors concluded that the “pulse action of the current on zinc single crystals during creep is accompanied by a considerable increase in the creep rate and by a discontinuous increase in deformation.” Conrad reported that electric and magnetic fields can often have a significant effect on the plastic deformation of metals and ceramics.42 Conrad performed experiments on the effects of external DC
___________________
32 Ibid.
33 O. Troitskii and V. Likhtman, 1963, “The Effect of the Anisotropy of Electron and γ Radiation on the Deformation of Zinc Single Crystals in the Brittle State,” Doklady Akademii Nauk SSSR 148:332–334.
34 L.B. Zuev, V.E. Gromov, and V.F. Kurilov, 1978, “Mobility of Dislocations in Zinc Single Crystals Under the Action of Electric Current Pulses,” Doklady Akademii Nauk SSSR 239(1):84–86 (in Russian).
35 L.B. Zuev, V.E. Gromov, and L.I. Gurevich, 1990, “The Effect of Electric Current Pulses on the Dislocation Mobility in Zinc Single Crystals,” Physica Status Solidi (a) 121(437).
36 V.I. Stashenko, O.A. Troitskii, and V.I. Spitsyn, 1983, “Action of Current Pulses on Zinc Single Crystals During Creep,” Physica Status Solidi (a) 79(549).
37 H. Conrad, 1998, “Some Effects of an Electric Field on the Plastic Deformation of Metals and Ceramics,” Materials Research Innovations 2(1):1–8, https://doi.org/10.1007/s100190050053.
38 L. Guan, G. Tang, and P.K. Chu, 2010, “Recent Advances and Challenges in Electroplastic Manufacturing Processing of Metals,” Journal of Materials Research 25(7):1215–1224.
39 A. Lahiri, P. Shantraj, and F. Roters, 2019, “Understanding the Mechanisms of Electroplasticity from a Crystal Plasticity Perspective,” Modelling and Simulation in Materials Science and Engineering 27:085006.
40 C. Rudolf, R. Goswami, W. Kang, and J. Thomas, 2021, “Effects of Electric Current on the Plastic Deformation Behavior of Pure Copper, Iron, and Titanium,” Acta Materialia 209:116776.
41 Stashenko et al., 1983, “Action of Current Pulses on Zinc Single Crystals During Creep.”
42 Conrad, 1998, “Some Effects of an Electric Field on the Plastic Deformation of Metals and Ceramics.”
electric fields during the superplastic deformation of a 7475 aluminum alloy. According to the author, the surface charge reduced the flow stress by 10–20 percent.43
Kim et al.44 discussed the origins of electroplasticity in metallic materials. They reported that, in a system that includes a grain boundary (i.e., general defect in polycrystalline metallic materials), charge imbalances near defects would “drastically weaken atomic bonding under electric current.” The authors also note, based on tests on magnesium and aluminum alloys, that “the weakening of atomic bonding was confirmed by measuring the elastic modulus under electric current, which is inherently related to the atomic bonding strength.” Xu et al.45 performed tensile tests on dog-bone specimens made with a Mg-3Al-1Sn-1Zn alloy. Current pulses with peak current densities of 0, 20, and 30 A/mm2 were applied to the specimens during tests. Based on stress-strain diagrams, the authors reported a drop in flow stress and increased fracture strain as the current densities increased. They reported that significant dynamic recrystallization developed with a current density of 30 A/mm2.
The flow of electrons in the cables caused increased temperature due to Joule heating. Stashenko et al.46 note that the “skin effect pressing a high-frequency current back to the outer areas of the sample … may result in an overheating of the surface.” Although creep is highly dependent on temperature, the Joule effect and the temperature rise due to the skin effect may not be significant compared to the potential electroplastic effects.
___________________
43 Ibid.
44 M.J. Kim, S. Yoon, S. Park, H.-J. Jeong, et al., 2020, “Elucidating the Origin of Electroplasticity in Metallic Materials,” Applied Materials Today 21, https://doi.org/10.1016/j.apmt.2020.100874.
45 H. Xu, Y.-J. Zou, Y. Huang, P.-K. Ma, Z.-P. Guo, Y. Zhou, and Y.-P. Wang, 2021, “Enhanced Electroplasticity Through Room-Temperature Dynamic Recrystallization in a Mg-3Al-1Sn-1Zn Alloy,” Materials 14(3739), https://doi.org/10.3390/ma14133739.
46 Stashenko et al., 1983, “Action of Current Pulses on Zinc Single Crystals During Creep.”
