machines

Review
Failure of Threaded Connections: A Literature Review
Dario Croccolo , Massimiliano De Agostinis * , Stefano Fini , Mattia Mele , Giorgio Olmi ,
Chiara Scapecchi and Muhammad Hassaan Bin Tariq

Department of Industrial Engineering (DIN), University of Bologna, 40136 Bologna, Italy

* Correspondence: [email protected]

Abstract: Threaded fasteners are vastly used in the industry due to ease of mounting and dismounting
and flexibility of design. Nonetheless, several researchers indicate that most failures recorded on
nearly any kind of machinery and vehicles are initiated at fasteners. This review paper aims at
summarising the failures involving threaded fasteners available in the literature. The analysis
involves categorization of the failures according to the root cause, e.g., incorrect assembly, overload,
fatigue, preload loss during operation, among others. The purpose of this review paper is to give
a systematic glance of the topic to the scientific and technical community, to correctly orient future
investigations on the broad topic of threaded fasteners.

Keywords: bolts; fasteners; failure; loosening; fatigue

1. Introduction
The joining of mechanical parts plays a key role in all industrial fields, such as
aerospace, automotive, machine tools and others. The failure of a joining mechanism
(permanent as well as non-permanent) can easily result in the failure of the whole ma-
chinery [1]. Threaded fasteners are one of the most widespread non-permanent joining
techniques due to ease of assembly and removal [2]: unreliability of threaded fasteners can
be a threat to the integrity of the whole structure. Failure of threaded fasteners can cause se-
Citation: Croccolo, D.; De Agostinis, rious damages to the machinery, sometimes even resulting in catastrophic failures including
M.; Fini, S.; Mele, M.; Olmi, G.; personal injuries, apart from financial losses [3]. Numerous failures that have been reported
Scapecchi, C.; Tariq, M.H.B. Failure of were initiated at threaded fasteners. The 911 GT3 automobiles manufactured by Porsche
Threaded Connections: A Literature were catching on fire, which led to Porsche recalling 785 cars for an engine replacement.
Review. Machines 2023, 11, 212. The reason behind that event was the loosening of an engine bolt, which allowed oil to leak
https://doi.org/10.3390/ into the exhaust, where it ignited [4]. Similarly, in 2014 GM motors recalled 8.4 million cars
machines11020212 because of problems including faulty fasteners [4]. Recently, Tesla Motors (Beijing, China)
Academic Editor: Gang Chen
Co., Ltd. found a problem in the steering system of X series automobiles due to which they
recalled more than three thousand cars produced from 2016 to 2020. The bolts joining the
Received: 31 December 2022 motor and housing of the steering system underwent corrosion when subjected to highly
Revised: 24 January 2023 corrosive environments. This problem may arise in areas where chlorine salt (corrosive
Accepted: 26 January 2023 environment) is sprayed on the roads to remove the snow [5]. Thus, this number gives an
Published: 2 February 2023
idea of how minor negligence in bolts can result in huge financial and, sometimes, human
losses because these failures can in turn lead to road accidents. Secondly, this problem does
not involve only the automobile industry: for example, a failure of bolts in a lower marine
riser package in the Gulf of Mexico resulted in leakages. After identification, 10,000 bolts
Copyright: © 2023 by the authors.
Licensee MDPI, Basel, Switzerland.
were replaced which also resulted in disruption of activities for some duration. The bolt
This article is an open access article
failure was attributed to hydrogen-induced stress corrosion cracking (HISCC) which was
distributed under the terms and
generated by an inadequate coating process [6]. Therefore, special consideration must
conditions of the Creative Commons be given to the several types of failure and a framework must be provided on how to
Attribution (CC BY) license (https:// tackle these various failures to update the bolt codes and standards, as well as the design
creativecommons.org/licenses/by/ procedures, in order to minimize the losses.
4.0/).

Machines 2023, 11, 212. https://doi.org/10.3390/machines11020212 https://www.mdpi.com/journal/machines

Machines 2023, 11, 212 2 of 41

There can be several types of failures associated with threaded fasteners. Some
types of failures can gradually and temporarily decrease the strength of the fasteners
whereas others can lead the bolts to sudden breakage, causing severe damage to the entire
system. Preload loss can occur due to several types of loadings on threaded fasteners.
It can lead to the loosening of the joint, which can combine with fatigue in the case of
vibrational loading to accelerate the failure [7]. Like loosening and fatigue, improper design
parameters or inappropriate material processing can also cause similar fastener failures [8].
Stress corrosion cracking is the type of failure in which fasteners fail abruptly in a brittle
manner. Therefore, prevention of these failures is crucial and only possible when the
mechanism behind the failure is known. Researchers have put forth efforts in this regard
by experimentally testing fasteners under different conditions and by setting up numerical
as well as analytical models to predict their behaviour.
Considering the large amount of information regarding the failure of threaded joints
available in the literature, the present contribution aims at systematizing the data in the
framework of a classification based on the root cause of failure presented by each of the
analysed papers and technical reports. In light of their research and industrial expertise
on the topic, the authors deem this classification useful for both researchers and design
engineers to understand the possible threats related to the application of threaded joints
in their specific context, as well as to highlight possible areas of improvement of the
knowledge on the topic, which might help to orient future research efforts.

2. Loosening and Preload Loss (Dynamic or Static)

Loosening is a process of rotation of the bolt in a direction opposite to the tightening.
This can cause a loss of preload (tension created in the screw shank upon tightening)
which can cause failures in the machinery [9]. For instance, Olmi [10] investigated the
crack generated in a geartrain housing of an asphalt milling machine. The experimental
simulation of the entire process with the assembly and the possible external loads showed
that there was a stress increase due to the load hysteresis. The stress peak was present
especially at the curvature where the crack originated. This load hysteresis was made
possible due to the vibrational loading by the engine, which resulted in the self-loosening
of the bolts. Therefore, he suggested design improvements along with the use of dampers
to prevent the vibrational loading generated by the engine. Another preload loss due
to dynamic loading and reported in the literature occurred in a pinion shaft for wheel
loaders [10]. Croccolo et al. [11] investigated the fracture of the pinion shaft at its threaded
end. A ring nut was used to preload the angular roller bearings. It was shown that the
preload loss that took place under the nominal duty cycle is around 15 kN, due to which
loosening starts occurring. This loosening results in the crack of the weakest portion of the
assembly, i.e., the root of a sharp notch next to the threaded portion where the ring nut
engages. The suggestion for maintaining the preload given was to use either an adhesive
threadlocker or a mechanical locking device. Otherwise, the design of the assembly should
be altered to lessen the amount of the loading, which includes using more compliant sleeves
or even removing the inner sleeve [11].
Preload loss of threaded fasteners is also affected by the types of materials in use.
For instance, Den Otter and Maljaars [12] reported that when stainless-steel bolts are
used in aluminium-plated slip-resistant connections, significantly greater preload loss is
observed as compared to the carbon steel bolts as specified in ISO 898-1. This preload
loss is due to several reasons, among which viscoplastic behaviour, plate contraction and
drop in the temperature are mentioned by the authors. The viscoplastic behaviour includes
creep and relaxation phenomena. Therefore, it was suggested to use a higher safety factor
(1.3–1.5 instead of 1.25 provided in EN 1999-1-1) depending on connection configuration to
compensate for the increased preload loss due to material combination in the ultimate limit
state [12]. It is worth remarking that a higher factor of safety determines an increase in the
weight and cost of the engineering structure [13]. Therefore, if these aspects are crucial, the
safety factor should be minimized by taking into account all the available information such
Machines 2023, 11, 212 3 of 41

as thread friction, rough surface finishing and the actual distance between the bolt axis and
the edge of members [14].
Moreover, the initial preload application can also affect the loosening of the bolted
connections. Bhattacharya et al. [15] studied the effects of initial preload application on
M16 steel bolts that loosening is minimum when the initial preload is greater than or equal
to 12 kN, as shown in Table 1. This might be due to the pronounced deformation of the
points of contact with greater force values, which enable them to offer more friction [15].

Table 1. Loss of preload affected using different amounts of initial preload application for M16 steel
bolt [15].

Initial Clamping Force (kN) Percentage of Loosening

8 16
9 6.4
10 6.0
11 5.9
12 3.4
13 3.0
14 2.5
15 2.5
16 2.7
17 3.0
18 3.6

2.1. Theories of Loosening Process

Various theories have been put forward to explain the process of loosening. The
first theory on the rotation mechanism in bolts was coined by Sakai [16] in 1978. He
reported that loosening occurs when the torque for bolt underhead slip is greater than
the loosening torque for thread surface slip and lesser than the tightening torque required
by the threaded portion. It was assumed in the theory that the nut is fixed; hence, the
condition for loosening was based only on the torques on the bolt surfaces. In the classical
theory used by Junker [17] (also known as the “complete slippage theory”), loosening is
said to occur when frictional forces on the threaded surfaces are completely overcome.
Yamamoto and Kasei [18] proposed a two-stage theory, where the first stage represents
slippage on bolt-nut thread surfaces producing torsion on the bolt shank. In the second
stage, the rotation of the bearing surface occurs on a moving plate, which results in a release
of torsional deformation and produces loosening. In Junker’s theory, the bolt loosens first;
however, in Yamamoto’s theory, the nut loosens first [18]. In contrast to the classical theory,
Pai and Hess [19] proposed the “local slippage accumulation theory”. According to such a
theory, complete slippage is not necessary to initiate loosening. Loosening starts when local
areas of slippage accumulate together over the number of vibration cycles. Recently Li
et al. [20] extended Sakai’s theory of self-loosening to include free nuts and free bolts under
rotary vibrational loading, which agreed well with the experimental evidence. According
to their theory, local slippage does not account for loosening, as opposed to the Pai and
Hess theory. In Pai and Hess’s theory, the transverse load can introduce eccentricity, which
in turn allows the piling up of local slippage. However, Li et al. [20] introduced a theory for
rotational loading in which there is no eccentricity and hence does not allow accumulation
of local slippage. These theories are summarised in Table 2.
Machines 2023, 11, 212 4 of 41

Table 2. Theories of loosening mechanisms presented by different researchers.

Theory Reference
Sakai’s fixed nut and free bolt head theory [16]
Junker’s complete slippage theory [17]
Two-stage theory by Yamamoto and Kasei [18]
Local slippage theory [19]
Theory for rotational loading [20]

2.2. Analytical Modelling

Loss of preload can occur with or without external excitation. This phenomenon of
loss of preload has been studied from both standpoints in the past. Bretl and Cook [21]
performed a static numerical analysis of the loading distribution over the threads during
bolt tightening. Moreover, for the calculation of thread force, Miller et al. [22] established
an equivalent spring model for fastening of the bolt and the nut. This approach is relatively
simple and immediate. Miller’s model was later improved by Daadbin [23] by adding
stiffness and damping, which he claimed to be influential factors for load transmission.
Further investigation into the problem was performed by Daadbin and Chow [24], who
examined the impact of the loading scheme on the preload loss. However, they neglected
the deformation of the thread and modelled only one thread, because of which their model
was unrealistic. Nevertheless, they concluded that the preload loss is less for the fine
pitch, as it leads to a greater coefficient of friction and duration of applied force [24].
Moreover, an exponential expression for the stiffness of the members of bolted connections
was developed using the numerical modelling by Wileman et al. [14]. Additionally, Chen
et al. [1] made a significant improvement in Miller’s model by considering a multiple-thread
system and adding the threads and bolt shank deformations. They modelled both the static
and dynamic loading of the bolts and also determined if the bolt is exceeding the elastic
limit by excessive initial preload. They calculated the effects of various aspects including
the lead angle, material type, and external excitation parameters using the developed
models. They concluded that a larger preload results in the extended service life of a bolted
connection, whereas the trend is opposite for the lead angle of the threads and frequency of
excitation force. They demonstrated the use of the model in the optimization of the design
of a missile tail wing. Using the model, it was shown that preload loss, e.g., was decreased
by decreasing the lead angle of the threads [1].
The various kind of loadings on the bolted joints can result in the loosening of the
bolted joints. Work has been done to study the loosening process under different loading
conditions such as axial, transverse, rotational vibration, impact loading and thermal
loadings. These different types of loadings are discussed in the preceding sections.

2.3. Vibrational Loading

The majority of the vibrational loading on the threaded fasteners results in the loos-
ening phenomenon. As an example, Casanova [25] studied the fracture of bolts joining
the draft tube of a Francis turbine. The failure originated due to the presence of extremely
high vibrational loading. The assembly failed in fatigue due to the loosening of the bolts
connecting the draft tube with the lower flange. The preload loss characteristics of bolts
must be increased, or the lower assembly should be designed such that preload loss is
prevented in the bolts working under conditions of vibrational loading [25].
Housari and Nassar [26,27] developed a linear mathematical model to represent
the self-loosening behaviour of bolts subjected to transverse vibrational excitation and
validated it experimentally. Later, Yang and Nassar [28,29] developed more accurate, non-
linear models for representing the bolt self-loosening subjected to transverse vibrational
excitation. In their models, they considered the beam-bending equations to evaluate the bolt
deflection and angle of rotation in a direction opposite to the tightening when it is subjected
Machines 2023, 11, x FOR PEER REVIEW

subjected to external excitation. They considered a non-linear pressure distributi

Machines 2023, 11, 212 5 of 41
the bolt head and on the thread surface, yet neglected the effect of nut thread
Recently, Nassar and Yang [30] improved the same mathematical model to encom
toeffect
externalofexcitation.
the nut-thread geometry
They considered in the thread-friction
a non-linear pressure distributionanalysis
underof thethe
boltbolt–nu
which
head and came
on the out to be
thread in good
surface, agreement
yet neglected with of
the effect thenut
experimental
threads [28,29].results.
Recently,Howeve
Nassar and Yang [30] improved the same mathematical model to encompass
ical modelling still requires many parameters, such as the plasticity of the threa the effect of the
nut-thread geometry in the thread-friction analysis of the bolt–nut system, which came out
taken into account for a realistic approach, which makes it much more comput
to be in good agreement with the experimental results. However, analytical modelling still
requires many [31].
expensive Therefore,
parameters, such asmany researchers
the plasticity have studied
of the threads, theinto
to be taken loosening response
account for
joints subjected
a realistic to vibration
approach, which experimentally
makes it much as well as
more computationally numerically.
expensive [31]. Therefore,
many researchers have studied
Experimentally, the loosening
Goodier response of [32]
and Sweeney boltedstudied
joints subjected to vibration
the axial vibrational lo
experimentally as well as numerically.
1945 at Cornell University using microscopes to measure the nut rotations. Follow
Experimentally, Goodier and Sweeney [32] studied the axial vibrational loading in
experimentation,
1945 at Cornell University different researchers
using microscopes improved
to measure their
the nut apparatus
rotations. Followingandtheir
studied th
of various parameters
experimentation, on the loosening
different researchers improved their [33,34]. In 1969,
apparatus an apparatus
and studied the effectsfor
of the ap
of vibrational loading to the bolted connections was developed by Junker [17
various parameters on the loosening [33,34]. In 1969, an apparatus for the application of
vibrational
opened loading
a new to the bolted
phase connections
of research in was
thedeveloped
loosening byofJunker [17],connections.
bolted which opened Junker
a new phase of research in the loosening of bolted connections. Junker used an eccentric
eccentric mechanism to apply the vibrational loading to the bolted connection
mechanism to apply the vibrational loading to the bolted connection connections, the basic
tions, theofbasic
mechanism whichmechanism of which
is shown in Figure 1. is shown in Figure 1.

Figure 1. Schematic diagram of Junker apparatus used to apply vibrational loading to the bolted
Figure 1. Schematic diagram of Junker apparatus used to apply vibrational loading
connections.
to
connections.
His apparatus is still used by researchers with some modifications [27,35] and it
also laidHis
the apparatus
foundations is forstill
current
used standardized
by researcherstestingwith
of fasteners
some against vibrational
modifications [27,35] a
loosening [36]. Another German engineer, Dinger [37], developed a more sophisticated
laid the foundations for current standardized testing of fasteners against vibratio
machine for applying transverse and rotational vibrational excitations to bolted connections.
ening
Using his[36]. Another
apparatus, Germanloading
the combined engineer, Dingerand
of translation [37], developed
rotation can alsoabemore
appliedsophistic
tochine for connection.
the bolted applying transverse andthat
It has been found rotational vibrational
self-loosening excitations
could occur below the to bolted con
critical
kinematic parameters used in pure rotational or pure translational loading.
Using his apparatus, the combined loading of translation and rotation can also b
Dinger
to the [38] also
bolted used a complete
connection. It hasFEAbeen model,
found whose
that working principlecould
self-loosening is shown in below
occur
Figure 2, to replicate the bolt loosening phenomenon. He used the local slippage theory
toical
findkinematic parameters
out the critical condition forused
the in pure rotational
rotational loosening. orThepure translational
results of FEA were loading.
in
Dinger [38] also used a complete
agreement with the experimental analysis performed. FEA model, whose working principle is
Figure
Chen 2,
et to replicate
al. [39] the FEA
performed bolt for
loosening
studying phenomenon.
the behaviour of theHe bolts
usedsubjected
the local to slippag
transverse cyclic loading. The authors concluded from the simulations for
to find out the critical condition for the rotational loosening. The results of FEA tightening and
loosening that a creep slip phenomenon is present, which permits the loosening of the bolt
agreement with the experimental analysis performed.
even with some of the stuck contact facets.
Machines 2023, 11, 212 6 of 41

Figure 2. Finite element model of the bolted connection using the localized slippage theory [38].

Chen et al. [39] performed FEA for studying the behaviour of the bolts subjected to
transverse cyclic loading. The authors concluded from the simulations for tightening and
loosening that a creep slip phenomenon is present, which permits the loosening of the bol
even with some of the stuck contact facets.
Moreover, in order to study the loosening response towards random vibration in PCB
Figure 2. Finite element model of the bolted connection using the localized slippage theory [38].
bolts, Du et al. [40] developed an experimental configuration to provide triaxial random
vibrations
Moreover,using a triaxial
in order vibrostand
to study as shown
the loosening in Figure
response towards3.random
The base of theinsetup
vibration PCB is fixed
denoted
bolts, Du etbyal.position 5, and an
[40] developed theexperimental
PCB (position 4) is mounted
configuration on thetriaxial
to provide base. random
Positions 3, 6, 7
vibrations using a triaxial
and 8 correspond vibrostand
to triaxial as shown in Figure
accelerometers. 3. The
Positions base9ofare
2 and the relevant
setup is fixed,
to strain ro
denoted by position 5, and the PCB (position 4) is mounted on the base.
settes, and position 1 refers to the M10 UNC thread bolt. The authors also performed Positions 3, 6, 7 FE
and 8 correspond to triaxial accelerometers. Positions 2 and 9 are relevant
analyses for random vibrations and found a good agreement between the experimenta to strain rosettes,
and position 1 refers to the M10 UNC thread bolt. The authors also performed FE analyses
tion and the analysis. On the basis of their analysis, they proposed a three-stage criterium
for random vibrations and found a good agreement between the experimentation and
(stable,
the transition,
analysis. loosen
On the basis stages)
of their for the
analysis, self-loosening
they of the randomly
proposed a three-stage criteriumvibrating
(stable, speci
men [40].
transition, loosen stages) for the self-loosening of the randomly vibrating specimen [40].

Figure3.3.Experimental
Figure Experimentalapparatus for random
apparatus vibration
for random of the of
vibration bolted connections
the bolted [40].
connections [40].
Machines 2023, 11, 212 7 of 41

Table 3 enlists the various mathematical models as well as experimental techniques

used to investigate the self-loosening phenomenon.

Table 3. Review of literature on self-loosening due to vibrational loading.

Mathematical Models Reference

Linear self-loosening mathematical model [26,27]
Non-linear self-loosening mathematical model [28,29]
Incorporation of nut thread effect in a non-linear self-loosening model [30]
Experimental Analysis Reference
Use of microscopes to measure nut rotation [32]
Junker’s apparatus for transverse excitation [17]
Dinger’s apparatus for transverse and rotational excitation [37]
Experimental configuration for PCB subjected to random vibrations [40]

2.4. Impact Loading

Similarly, impact testing experimentation was performed initially by Baubles et al. [41].
They applied vibrations at excitation frequencies of 30 Hz and 60 Hz to the fixture, by means
of a universal fatigue testing machine and electromagnetic shaker, respectively. Zhang
et al. [42] developed an experimental apparatus to evaluate the effects of clamp length and
loading direction of bolted joints on their self-loosening behaviour. The apparatus consists
of a load cell for the measurement of clamp force, an angle transducer which measures the
rotation of the screw tip, and an extensometer, which picks up the axial displacement of the
joint. The line of action of the force was kept along the contact surfaces of the two plates in
order to prevent any undesired bending moments coincident with the line of action of the
screw force. Pin joints were used to attach the movable and fixed plates to the two jaws of
the universal testing machine, which enabled easy force calculations.

2.5. Thermal Loading

Miller et al. [43] performed experimental and numerical investigation on bolts applied
to a c-beam structure. They used a shaker with an amplifier for applying an external
excitation to the c-beam structure. Eraliev et al. [44] developed an experimental setup
to study the loosening behaviour of bolted connections under temperature changes. To
study the phenomenon, they used two steel plates of 30 mm thickness. For preload change
measurement, they used a washer-type compression load cell, while the temperature was
monitored by means of a K-type thermocouple. A laser sensor was used to measure the
movement of a thin plate attached to the nut, thus calculating the rotation of the nut. Heat
cartridges as shown in Figure 4 were used to heat up the specimen. They found a direct
proportionality between temperature increment and bolt preload loss, with preload being
greater in the 1st cycle as compared to the remaining cycles.
All the examples above indicate that the phenomenon of loosening must be considered
by engineers for designing a bolted connection with a particular focus on the type of
loading, materials and temperature changes that can occur.
Machines
Machines2023,
2023,11,
11, x212
FOR PEER REVIEW 8 of 41 8 of 4

Figure
Figure 4. 4. Experimental
Experimental setup
setup to study
to study self-loosening
self-loosening due to due to temperature
temperature changes [44].
changes [44].

3. Incorrect Assembly
All the examples above indicate that the phenomenon of loosening must be consid
eredImproper assembly
by engineers specifications,
for designing wrongconnection
a bolted execution or a combination
with a particularoffocus
both often
on the type o
play a critical role in the failures of bolted connections. Casanova and Mantilla [45] showed
loading, materials and temperature changes that can occur.
that the application of an extremely low preload compared to the design specification
(around 15% of the desired value, i.e., a coefficient of utilization of the screw ν = 0.64 w.r.t
3. Incorrect
the yield pointAssembly
of the material, as reported by the authors) during the assembly process
Improper
was a possible assembly
cause of failurespecifications,
in a bolted flange wrong execution
connecting or a combination
a turbine of both ofte
with an alternator
shaft. Such low preload levels, even in the absence of further loosening
play a critical role in the failures of bolted connections. Casanova and Mantilla [45 events—actually
quite
showedlikelythat
to happen under that of
the application condition—notoriously
an extremely low preload lead to increased
compared load
tofluctuation
the design specif
on the screw for given external fatigue loads. This occurrence may easily lead the screw to
cation (around 15% of the desired value, i.e., a coefficient of utilization of the screw ν
fatigue failure, as documented by the authors [45]. Similarly, Fonte et al. [46] investigated
0.64 w.r.t the yield point of the material, as reported by the authors) during the assembl
the failure of bolts connected to a marine main engine. The reason for the final failure
process
was was
fatigue, a possible
as shown by thecause of failure
fracture surfacein a bolted flange
morphology. connecting
Nonetheless, a turbine
the roots of the with a
alternator
thread showed shaft. Such loworpreload
no corrosion levels, even
poor machining. in the absence
The fatigue of further
failure most probablyloosening
occurred events—
actually
due to the quite likelyoftoa happen
application under20%
screw preload thatlower
condition—notoriously
than that prescribed inlead to increased loa
the assembly
fluctuation
procedure onAnother
[46]. the screw for given
failure due to external
an improper fatigue loads.
preload This occurrence
application procedure may
(pluseasily lea
poor
the design
screw of to the joint)failure,
fatigue was reported by Abdul-Jawwad
as documented by the et al. [47],[45].
authors who Similarly,
studied the Fonte
failureet al. [46
ofinvestigated
bolts attachedthe to afailure
heavy of fuelbolts
oil meter. Fatiguetofailure
connected took place
a marine main atengine.
the thread
Therunout
reason for th
of several bolts, at a sector of the flange of a pressure vessel made of two semi-spherical
final failure was fatigue, as shown by the fracture surface morphology. Nonetheless, th
parts. The authors concluded that the failure originated due to an incorrect tightening
roots of the thread showed no corrosion or poor machining. The fatigue failure most prob
procedure (the technicians probably did not follow well-established sequences, such as the
ably
star occurred
pattern). Suchdue to the procedure
a wrong application ledoftoathe
screw preload
bending 20%screws
of some lower that,
thancombined
that prescribed i
the the
with assembly procedure
loads generated [46].
by the Another
duty failure due
cycle, brought abouttofatigue
an improper
failure ofpreload application
some bolts. The pro
cedure (plus poor design of the joint) was reported by Abdul-Jawwad
authors put forth a possible explanation for the wrong procedure adopted in the field, i.e., et al. [47], wh
studied
the the for
difficulty failure of bolts attached
the technicians to reachtosome a heavy
boltsfuel
of theoilassembly.
meter. Fatigue
On topfailure
of that,took
the place a
the thread runout of several bolts, at a sector of the flange of a pressure vessel made o
two semi-spherical parts. The authors concluded that the failure originated due to an in
correct tightening procedure (the technicians probably did not follow well-established se
quences, such as the star pattern). Such a wrong procedure led to the bending of som
Machines 2023, 11, 212 9 of 41

joint was designed such that the thread runout coincided with the contact surface of the
flanges, which is usually deprecated [48].
As a matter of fact, the process of bolt fastening to achieve the required preload may
turn out to be all but straightforward. In bolted flanged joints, the amount of preload
is affected by the surrounding bolts due to the elastic interaction of each bolt with the
joining material. Due to this, in the pressure vessels and piping industry, special tightening
sequences have been developed to achieve uniform initial preload on multi-bolted joints.
One of those methods is the elastic interaction coefficient method introduced by Van
Campen [49]. It was further developed by Bibel [50], according to whom interaction elastic
coefficients are calculated from experimental data and are then used to calculate the final
preload acting on the bolts. However, this method cannot be applied to non-linear systems.
Nassar et al. [51] introduced another method, known as the inverse square method. In this
method, all the bolts are fastened to the joint and then unfastened. During dismounting,
the analysis of loads from the final to the initial loading state is done and the loads to be
applied are measured. This method requires the initial unloading of all the bolts, which is
one of its limitations [51]. Abasolo et al. [52] developed a metamodel for bolting sequences
using the aforementioned methodologies. The optimization was done in Matlab at a very
low computational cost, which could be used to optimize the bolts preload in one or two
passes. Zheng et al. [53] developed another tightening sequence based on elastic interaction
coefficient and gasket creep relaxation. The algorithm was used to reduce the dispersion of
the clamping force. The dispersion of clamping force was demonstrated to drop when two
passes were applied using the proposed algorithm with the nut factor from the first pass as
an input.
As a final remark, multi-bolted joints are characterized by additional challenges with
respect to single bolt joints: due to that, the tightening sequence plays a key role in the
achievement of a correct initial force on each bolt. Without the correct sequence, the
achieved initial preload would diverge from the required value, eventually leading to the
failure of the joint as documented by the papers cited above.

4. Improper Design Parameters

The design of different bolt features such as thread surfaces, thread root radii and
underhead bearing surfaces must be considered carefully. The design of bolted joints must
also consider the number of bolts for various loading conditions of the application and the
required initial preload for the service life. Sharp corners between the head and shank and
at the thread root can be a source of high-stress concentration, which can lead to failure;
hence, they must be minimized. Shiba et al. [54] assessed a Hoffmann–Vidal external
fixation apparatus subjected to fatigue loading. The fatigue failure occurred at the point
where the post diameter was reduced to fit into the pin bracket, which was a region of
high stress concentration due to the change in diameter. It was not possible in this case to
change the material of the apparatus. The appropriate design suggestion, therefore, was
to press-fit the post into the pin bracket without a reduction in the diameter to avoid the
stress concentrations at that point.
Similarly, Milan et al. [3] reported the failure of steel bolts used in speed reducer hous-
ings after a few hours of service. They performed metallurgical and mechanical analyses of
the failed specimen. The root-cause of failure turned out to be the non-uniformity of the
roots of the threads, which resulted in high stress concentration; moreover, surface defects
and low toughness of the material contributed to the final failure of the bolts [3]. Curtis [55]
analysed the failure of the joint between an impeller blade and the rotating assembly. The
possible reasons for the failure included the sharp radius at the head-to-shank transition,
the sharp thread root radius and an inaccurate application of preload. Additionally, the bolt
thread shape was non-standardized, which must have resulted in the incomplete fitting of
the male and female threads and hence led to the occurrence of fretting, as indicated by the
microscopic images.
Machines 2023, 11, 212 10 of 41

All the above poor design considerations led to crack initiation and failure of the bolted
connection. The authors put forward some suggestions to prevent the failure, including
(i) adopting rolled threads to obtain an additional compressive stress layer, (ii) improving
the accuracy of the tightening torque measurement to avoid overtightening of the bolts,
(iii) using bolts with a larger head-to-shank radius, and (iv) adopting rounded thread roots
and standardized thread shapes to provide an improved matching between the thread
geometries [55].
Moreover, the thread shape of the fasteners affects the stress distribution during
assembly and service. Eraslan and İnan [56] studied the effects of the thread shape of
titanium screw implants in the bone environment. They studied the four thread shapes
including V thread, buttress, reverse buttress, and square threads. The von Mises stress
distribution was similar for all the thread types. However, the distribution of compressive
stresses was different. The maximum value of compressive stresses was 18 MPa in all the
thread shapes. Reverse buttress thread had the maximum intensity area covered by the
compressive stresses compared to the other thread types [56]. Similarly, Alemayehu and
Jeng [57] investigated the stress distribution of dental implants having titanium grade 4.
They found that the employment of different thread shapes resulted in different stress
distributions in the fasteners. Maximum von Mises stresses occurring in the implant region
of the prosthesis for the static and dynamic loading are shown in Table 4. It can be seen
clearly from the table that stress distribution changes significantly by changing the thread
shape. Under dynamic loading, triangular threads have almost 68% more maximum von
Mises stress than square threads.

Table 4. Maximum von Mises stress (MPa) in the implant region of the implant under static, quasi-
static and dynamic loading [57].

Thread Type Static Loading Quasi-Static Loading Dynamic Loading

Square 121.04 126.49 159.77
Buttress (CF) 173.24 182.33 239.41
Reverse buttress (CF) 205.79 214.42 298.60
Trapezoidal 237.23 247.06 358.69
Triangular 313.81 313.81 496.48

Similarly, the friction coefficients of the bolts and their lubrication state must be given
due consideration while designing the joints. Stevenson et al. [58] reported the changes in
frictional characteristics of the bolts during retightening to be a potential cause of failure
of the bolts in a commercial jetliner. As recalled by the standards [59], slight changes in
the friction coefficients can result in a significant change in preload for given tightening
torque. Furthermore, during metallurgical and mechanical property evaluation, no material
deficiency was found which could lead to failure. Hence, the authors suggested periodic
replacement of the fasteners with new ones after being used. Moreover, when the materials
of the bolts or nuts are changed, the value of the friction coefficients change, hence affecting
the design of the bolts drastically. Likewise, Croccolo et al. [60] performed the failure
analysis of bolted joints comprising aluminium alloy components used in front motorbike
suspensions and studied the friction coefficients of various process parameters of the parts.
The difference in friction coefficients indicated that applied torque must be matched to the
process by which the parts are treated (e.g., spray painting vs. anodization). Moreover, it
was suggested that during maintenance, with an increasing number of tightenings, the
tightening torque should be adjusted to obtain the same amount of preload force [60].
Croccolo et al. [61] also studied the behaviour of titanium bolts with aluminium underhead
bushings and steel nuts. They measured the torque and the friction coefficients for twenty
retightening cycles on dry bolts and also used different lubricants (Teflon-added oil and
ceramic paste). They reported that suitable lubrication is required when using titanium fas-
 (Teflon-added oil and ceramic paste). They reported that suitable lubrication is
when using titanium fasteners in combination with aluminium parts in order to
the desired level of preload and good stability thereof over repeated tightenings
the two lubrications, the ceramic paste showed a better performance in terms of
Machines 2023, 11, 212 11 of 41
of the preload (i.e., friction coefficients) over repeated tightenings, by protecting
tact surfaces from excessive wear. As a general statement, the use of proper lubri
necessary
teners for achieving
in combination the correct
with aluminium partspreload
in order to during
achieverepeated
the desiredtightenings
level of preload [61]. De A
et al.
and good [62] studied
stability the over
thereof friction coefficient
repeated in various
tightenings. Among tightening and retightening
the two lubrications, the
ceramic
bolts, paste showed
involving a better performance
high-strength gradeinboltsterms(12.9)
of stability of the preload (i.e.,
in combination withfriction
cast iron pa
coefficients) over repeated tightenings,
the fifth retightening, the valueby ofprotecting
the mean thefriction
contact surfaces
coefficientfrom for
excessive
the dry sur
wear. As a general statement, the use of proper lubrication is necessary for achieving the
played
correct greater
preload duringvariation
repeatedas compared
tightenings [61].toDethe case of
Agostinis lubrication
et al. [62] studiedby theceramic
friction paste
ver, the in
coefficient surfaces of the bolts
various tightening deteriorated
and retightening lessofwith
cycles bolts, the application
involving of ceramic
high-strength
shown
grade boltsin(12.9)
Figure 5. Han et al.
in combination with[63]
castaddressed
iron parts. the Afterissue of thread
the fifth galling
retightening, the in bolt
value of the mean friction coefficient for the dry surface displayed
through lubrication. They employed zinc oxide (ZnO) nanofluids for eliminating greater variation as t
compared to the case of lubrication by ceramic paste. Moreover, the surfaces of the bolts
lem of thread galling in threaded fasteners. However, the problem of agglome
deteriorated less with the application of ceramic paste, as shown in Figure 5. Han et al. [63]
nanofluids
addressed needs
the issue of to be addressed
thread by using
galling in bolted different
joints through surfactants
lubrication. Theytoemployed
enable them to
in bolt
zinc oxidethreads to prevent
(ZnO) nanofluids for excessive
eliminatingwear [63]. Kumar
the problem of thread etgalling
al. [64]in addressed
threaded the e
fasteners.
storage However,
conditions the on
problem of agglomeration
the frictional of nanofluids
characteristics of needs
bolted to joints.
be addressed
Theyby investiga
using different surfactants to enable them to be used in bolt threads to prevent excessive
Zn-flake coated threaded fasteners under cold (1 day, −20 °C) as well as hot an
wear [63]. Kumar et al. [64] addressed the effects of storage conditions on the frictional
conditions (2–30
characteristics of bolteddays, 40 They
joints. °C, 95% RH). They
investigated four showed a reduction
Zn-flake coated threaded infasteners
the friction coe
down
under coldto(1half
day,of−20 theC)initial
◦ as wellvalue.
as hot andIn cold
humid environments,
conditions (2–30this days,reduction
◦
40 C, 95% RH). could be a
to the
They presence
showed of free
a reduction water,
in the providing
friction coefficientslubrication
down to halfto of the underhead
the initial value. Inandcoldthread
environments,
In hot/humid thisconditions,
reduction could the be attributed
reason to the presence
for friction reductionof freemight
water, providing
be the formatio
lubrication to the underhead and thread surfaces. In hot/humid conditions, the reason
oxides. Among the two, cold storage turned out to be the most critical one, redu
for friction reduction might be the formation of zinc oxides. Among the two, cold storage
torque
turned outtotoeven
be theone-third
most criticalcompared
one, reducing to the tightening
the torque to evenofone-third
the reference
compared parts
to (no a
thetightening
the same preload level. Inparts
of the reference light(noofageing)
the above, the frictional
at the same preload level. behaviour
In light ofofthe the joint
above, the frictional behaviour of the joint must be given full consideration
given full consideration in the design phase: the state of lubrication, as well as th in the design
phase: the state of lubrication, as well as the nature of the coating applied to the fasteners,
of the coating applied to the fasteners, are of paramount importance in the defi
are of paramount importance in the definition of the torque-preload relationship. Therefore,
thedesigner
the torque-preload
must also take relationship. Therefore,
the storage conditions, thethe designer
number must re-tightenings
of expected also take the storag
tions,
and the number of
the environmental expected
conditions re-tightenings
in general into account and theprescribing
when environmental conditions in
the tightening
into account
torque when
for a specific prescribing the tightening torque for a specific application.
application.

Figure
Figure 5. Underhead
5. Underhead bearing
bearing surface
surface of the boltof the(a)bolt
after after (a)
tightening tightening
without without
lubrication, lubrication, (b
(b) tightening
ing with the ceramic paste
with the ceramic paste [62]. [62].

One more important design parameter are the loading conditions acting du
service of the bolts. Rezvani et al. [65] studied the behaviour of angle bracket con
under several types of loading conditions which they encountered in wall-to-fl
nections. For the uplift and in-plane shear loading conditions, a significant am
Machines 2023, 11, 212 12 of 41

One more important design parameter are the loading conditions acting during the
service of the bolts. Rezvani et al. [65] studied the behaviour of angle bracket connections
under several types of loading conditions which they encountered in wall-to-floor con-
nections. For the uplift and in-plane shear loading conditions, a significant amount of
plastic deformation occurred because the elastic limits of both the angle bracket and bolts
were exceeded. In the out-of-plane tension, there was a splitting of cross-laminated timber
(CLT) panels. Hence, incorrect knowledge of the loading carried by bolts during service
can hamper the performance of the joint.
Additionally, when designing bolted connections, the assumptions in the theoretical
models must be assessed with great care. Croccolo et al. [66] studied the behaviour of
screw clamps in front motorbike suspensions. They showed that the theoretical model
usually adopted by textbooks for the calculation of screw clamps (based on a simple beam
model) cannot catch the effect of geometrical irregularities. Therefore, by leveraging nu-
merical analyses they suggested adding some corrective parameters to the basic analytical
models in order to improve the accuracy of calculation on such specific kinds of bolted
connections [66].
When designing composite bolted joints, the ply orientation and number, type of
matrix and other structural parameters of the parts to be joined change the behaviour of
the joint under the loading conditions. Researchers have carried out parametric studies
on composite bolted joints and developed different failure criteria [67–70]. Among the
various possible failure modes, bearing failure is the phenomenon in which bolted joints
are designed to fail because of their progressive (thus, predictable) nature. Simplifying the
problem, many authors consider the bolted composite joint as a pin joint, neglecting stresses
other than in-plane stresses. Chang–Chang and Chang–Lessard formulations are among
the criteria for bolted composite joints [71,72]. Moreover, non-linear shear stress–strain
formulations have been applied in the past [73]. A more realistic picture of the phenomenon
can be represented by a 3D formulation which takes into account the 3D bearing failure in
a more sophisticated manner [74]. Moreover, out-of-plane stresses have been considered
by Hashin [75] to present a criterion for failure which has been applied by researchers
in bolted composite joints. Olmedo and Santiuste [76] presented a three-dimensional
failure criterion considering the out-of-plane stresses and non-linear shear stress–strain
relationship based on Chang–Chang and Chang–Lessard criteria. Finally, the hole diameter
is important in the definition of the strength of bolted composite joints. There are several
methods to determine the critical hole diameter, such as the method of progressive failure,
characteristic length or failure envelope method [77–79]. Among these, the failure envelope
method is cost-effective and timesaving, making it very appealing for real applications. In
the failure envelope method, it is assumed that failure strength and stress concentration
factor at the hole plate are linearly related to each other. This method was proposed by
Hart-Smith and gained popularity because it made it easy to predict the bearing and tensile
failure of double-lap joints having multiple bolts and quasi-isotropic lay-up [79]. This
method is also adopted by ASTM D 7248/D 7248M-08 standard [80]. In 2013, Liu et al. [81]
proposed a modified failure envelope method, which considered the effect of by-pass load
on the bearing failure contrary to the conventional. The proposed approach was applied
to two-bolt and four-bolt double-lap joints with a quasi-isotropic layout to predict failure.
The predicted failure was in good agreement with the experiments. Moreover, this method
can reduce the risk of tensile loading failure which is not considered by the conventional
method. Cheng et al. [82] used the failure envelope method for two- and three-bolted joints.
They concluded that the failure envelope method was neglecting the composite damage
effect. In order to address this problem, they proposed another method to predict failure
more accurately. On the contrary, the Monte Carlo method is also used by researchers
because of its accuracy, but it requires numerous samples [83,84]. Liu et al. used the failure
envelope method, Monte Carlo simulation and spring-based method to propose a new
method which covered the drawback of inaccuracy and required few samples [85].
Machines 2023, 11, x FOR PEER REVIEW 13

Machines 2023, 11, 212 5. Material Properties 13 of 41

Material properties affect the behaviour of bolted joints. Different types of mate
are used to design threaded fasteners with different responses to the various loading
5.ditions.
MaterialBolts
Properties
are manufactured through various processes to ensure the required lev
Material properties
performance throughoutaffect their
the behaviour
service of bolted
life. joints.treatments
Various Different types of materials
(specific to the bolt m
are used to design threaded fasteners with different responses to the various
rial) are performed to achieve the specified mechanical properties. Any deviations loading
conditions. Bolts are manufactured through various processes to ensure the required
the specific process may generate undesired characteristics, which may in turn lead t
level of performance throughout their service life. Various treatments (specific to the bolt
failure of
material) arethreaded
performed fasteners.
to achieveInthe
addition,
specifiedmaterial
mechanicaldefects also adversely
properties. Any deviationsaffect the a
cation
from theof the threaded
specific fasteners.
process may generate undesired characteristics, which may in turn lead
Firstly, an overview of theInmain
to the failure of threaded fasteners. materials
addition, materialused foralso
defects bolted jointsaffect
adversely is given.
the Then
application
main issues of the threaded
affecting fasteners.fasteners are covered in the following sections.
threaded
Firstly, an overview of the main materials used for bolted joints is given. Then, the
main issues affecting threaded fasteners are covered in the following sections.
5.1. Type of Material
5.1. Type of Material
Different types of materials are used in the bolted joints specific to certain app
Different types
tions. The stress and of materials are usedbehaviour
deformation in the boltedofjoints
thesespecific to certain
materials applications.
changes the overall qu
The stress and deformation behaviour of these materials changes the
of the bolted joints. Mostly, the bolts are made of steel alloys standardized overall quality ofin ISO 8
the bolted joints. Mostly, the bolts are made of steel alloys standardized in ISO 898-1 [86].
[86]. Lee et al. [87] studied the behaviour of high-carbon steel and demonstrated tha
Lee et al. [87] studied the behaviour of high-carbon steel and demonstrated that the man-
manufacturing
ufacturing processprocess
determinesdetermines
an increaseanin increase
the tensilein the tensile
strength of thestrength of the bolt if
bolt if compared
pared to the raw material. The stress–strain curves of the raw material
to the raw material. The stress–strain curves of the raw material and manufactured bolts, and manufact
bolts,by
found found
tensileby tensile
testing on testing on5583
an Instron an Instron 5583
mechanical mechanical
press, are shownpress, are 6.
in Figure shown in Figu

Figure
Figure 6. 6. Stress–strain
Stress–strain behaviour
behaviour of high-carbon
of high-carbon steel boltsteel bolt compared
compared to raw
to raw material [87].material [87].

Kodur et al. [88] studied the high-temperature behaviour of grade 8.8 high-strength
Kodur et al. [88] studied the high-temperature behaviour of grade 8.8 high-stre
bolts. The bolts M16 and M18 were made from a SAE 10B38 steel bar, whereas the M22
bolts. Themade
bolts were boltsfrom
M16 and10B21
a SAE M18 steel
werebar.madeThe from
residuala SAE 10B38 steel
stress–strain bar,during
response whereas the
bolts were
cooling made
is shown from7.aThe
in Figure SAE 10B21
first set of steel bar. and
bolts (M16 TheM18)
residual
retain stress–strain response du
most of the strength
◦
cooling
and is shown
stress–strain in Figure
response 7. at
they have The
roomfirst set of bolts
temperature when(M16
heatedand M18)
up to 500 retain
C, as is most o
obvious
strength from Figure
and 7a,b. However,
stress–strain the bolts
response they(M22)
haveof at
theroom
second set retain theirwhen
temperature properties
heated up t
up to 400 ◦ C, as shown in Figure 7c. In all cases, bolts reach the yield point and undergo
°C, as is obvious from Figure 7a,b. However, the bolts (M22) of the second set retain
plastic deformation before fracture.
properties up to 400 °C, as shown in Figure 7c. In all cases, bolts reach the yield poin
Similarly, Hu et al. [89] investigated the material response of stainless-steel bolts, A2–70
undergo plastic
and A4–70 (M20 × 120) deformation
when exposedbefore fracture.
to elevated temperatures. The tensile behaviour was
measured for the unheated specimen as well as during the cooling of the heated specimen,
as shown in Figure 8. It was observed that for the unheated specimen, A4–70 has almost
the same ultimate tensile strength, but greater yield stress as compared to A2–70.
Machines 2023,
Machines 11, 11,
2023, x FOR
212 PEER REVIEW 14 14 of 42
of 41

Figure 7. Residual stress–strain response of high-strength bolts subjected to high temperatures (a)
M16 (10B38), (b) M18 (10B38), (c) M22 (10B21) bolts [88].

Similarly, Hu et al. [89] investigated the material response of stainless-steel bolts, A2–
70 and A4–70 (M20 × 120) when exposed to elevated temperatures. The tensile behaviour
was measured
Figure forstress–strain
7. Residual the unheated specimen
response as well as during
of high-strength the cooling
bolts subjected of the
to high heated speci-
temperatures (a)
Figure 7. Residual stress–strain response of high-strength bolts subjected to high temperatures
men,(10B38),
M16 as shown in Figure
(b) M18 8. (c)
(10B38), It was
M22 observed that
(10B21) bolts for the unheated specimen, A4–70 has al-
[88].
(a) M16 (10B38), (b) M18 (10B38), (c) M22 (10B21) bolts [88].
most the same ultimate tensile strength, but greater yield stress as compared to A2–70.
Similarly, Hu et al. [89] investigated the material response of stainless-steel bolts, A2–
70 and A4–70 (M20 × 120) when exposed to elevated temperatures. The tensile behaviour
was measured for the unheated specimen as well as during the cooling of the heated speci-
men, as shown in Figure 8. It was observed that for the unheated specimen, A4–70 has al-
most the same ultimate tensile strength, but greater yield stress as compared to A2–70.

Figure 8. Stress–strain curves for (a) grade A2–70 and (b) grade A4–70 stainless-steel bolts exposed
to high temperatures [89].
Machines 2023, 11, x FOR PEER REVIEW 15 of 42
Machines 2023, 11, x FOR PEER REVIEW 15 of 42

Machines 2023, 11, 212 15 of 41

Figure
Figure8.8.Stress–strain
Stress–straincurves
curvesfor
for(a)
(a)grade
gradeA2–70
A2–70and
and(b)
(b)grade
gradeA4–70
A4–70stainless-steel
stainless-steelbolts
boltsexposed
exposed
totohigh temperatures [89].
high temperatures [89].

Banetetal.
Ban al. [90]tested
tested grade10.9
10.9 high-performance(HP) (HP) steelbolts.
bolts. Thematerial
material was
Ban et al.[90]
[90] testedgrade
grade 10.9high-performance
high-performance (HP)steelsteel bolts.The
The materialwaswas
developedtotowork
developed workadequately
adequatelyatathigh
highloads
loadswhile
whileproviding
providingresistance
resistancetotofire
fireas
as wellas
as
developed to work adequately at high loads while providing resistance to fire aswell well as
corrosion.They
corrosion. Theyperformed
performedtensile
tensiletesting
testingon
onM20
M20andandM22,
M22,asasshown
shownin inFigure
Figure9.9.Due
Duetoto
corrosion. They performed tensile testing on M20 and M22, as shown in Figure 9. Due to
the
the presence of the martensitic structure, the elastic-to-plastic transition is smooth, as is
thepresence
presenceofofthe themartensitic
martensiticstructure,
structure,the
theelastic-to-plastic
elastic-to-plastictransition
transitionisissmooth,
smooth,asasisis
visible
visible in the stress–strain curves.
visibleininthe
thestress–strain
stress–straincurves.
curves.

Figure
Figure9.9.Stress–strain
Stress–strainresponse
Stress–strain responseofofof
response (a)
(a)M20
(a)M20and
M20and(b)
and M22
(b)
(b) grade
M22
M22 10.9
grade
grade HP
10.9
10.9 HPbolts,
HP atataat
bolts,
bolts, atemperature ofof20
a temperature
temperature of
20
°C [90].
◦
°C [90].
20 C [90].

Fracture
Fracturemodels
modelswere
wereused
usedby byCai
Caietetal.
al.[91]
[91]totostudy
studythe
study themechanical
mechanicalbehaviour
mechanical behaviourof of
ASTM
ASTM A325
ASTM A325 bolts
A325bolts and
boltsand A572
andA572 Grade
A572Grade
Grade 50
50 50steel.
steel. Their
Their
steel. findings
findings
Their show
show
findings show a
a goodgood agreement
agreement
a good be-
between
agreement be-
tween
tweenexperimental
experimental
experimental results
results and
andthose
and those
results obtained
obtained
those by
by FEA.
obtained byFEA.
The The
FEA. Thetensile
tensile comparison
comparison
tensile between
between
comparison the
between
the measured
measured and and calculated
calculated stress–strain
stress–strain curve
curve of of
A325A325 bolts
bolts is is
shownshown
in
the measured and calculated stress–strain curve of A325 bolts is shown in Figure 10. in Figure
Figure 10. 10.

Figure
Figure10.
Figure 10.Stress–strain
10. Stress–straincurve
Stress–strain curveofof
curve ofA325
A325bolts
A325 boltsobtained
bolts obtainedby
obtained byexperimental
by experimentaltests
experimental testsand
tests andFEM
and FEMmodelling
FEM modelling[91].
modelling [91].
[91].

Titaniumhas
Titanium has beenusedused asaaboltbolt materialininmany
many sectorsdue due to itshighhigh strength-
Titanium hasbeenbeen usedas as a boltmaterial
material in manysectors
sectors duetotoits its highstrength-
strength-
to-weightratio,
to-weight ratio,biocompatibility
biocompatibilityand andhigh
highcorrosion
corrosionresistance.
resistance.LiLietetal.
al.[92]
[92]performed
performed a
to-weight ratio, biocompatibility and high corrosion resistance. Li et al. [92] performedaa
comparativestudy
comparative study betweenthe the titaniumalloy
alloy (Ti6Al4V)and and differentsteelsteel bolts.Their
Their
comparative studybetweenbetween thetitanium
titanium alloy(Ti6Al4V)
(Ti6Al4V) anddifferent
different steelbolts.
bolts. Their
findings
findings suggest that the ultimate tensile strength of titanium bolts is comparable to that of
findingssuggest
suggestthat
thatthe
theultimate
ultimatetensile
tensilestrength
strengthofoftitanium
titaniumbolts
boltsisiscomparable
comparabletotothat that
stainless-steel bolts
ofofstainless-steel A4–80.
stainless-steelbolts
boltsA4–80.
A4–80.
Aluminiumalso
Aluminium also givesthe the advantageof of weight reduction
reduction in machine frames.frames. The
Aluminium alsogives gives theadvantage
advantage ofweight
weight reductionininmachine
machine frames.The The
mechanical
mechanical performance
performance of aluminium
ofofaluminium bolted
bolted joints
joints was
was investigated
investigated by
by Adeoti
Adeoti etet al.[93]
al. [93]
mechanical performance aluminium bolted joints was investigated by
by means of tensile testing. The stress–strain response of a high-strength 6082-T6 alloy isAdeoti et al. [93]
shown in Figure 11. Their findings show that the aluminium’s tensile strength is much lower
than titanium or steel bolts, which determines a reduction in the performance of joints.
 Machines 2023, 11, x FOR PEER REVIEW 16 of 42

Machines 2023, 11, x FOR PEER REVIEW 16 of 42

by means of tensile testing. The stress–strain response of a high-strength 6082-T6 alloy is

shown in Figure 11. Their findings show that the aluminium’s tensile strength is much
Machines 2023, 11, 212 by means of tensile testing. The stress–strain response of a high-strength 6082-T6 alloy
16 of is
41
lower than titanium or steel bolts, which determines a reduction in the performance of
shown in Figure 11. Their findings show that the aluminium’s tensile strength is much
joints.
lower than titanium or steel bolts, which determines a reduction in the performance of
joints.

Figure 11.Stress–strain
Figure11. Stress–straincurve
curveofofhigh-strength
high-strengthaluminium
aluminium6082-T6
6082-T6[93].
[93].

FigureThe
11. different
The Stress–strain curve
differenttypes
types ofmaterials
high-strength
ofofmaterials used aluminium
usedininthe 6082-T6
theliterature are
literature [93].
aresummarized
summarizedininTable
Table5.5.

TableThe
Table different
5.5.Main
Main typestypes
types of materials
ofofmaterials
materialsused
usedforused
boltsin
bolts
for the
ininthe literature
thereviewed are summarized in Table 5.
reviewedliterature.
literature.

Serial
Table Number
Serial
5. MainNumber Material
types of materials used for bolts in of
MaterialtheBolt
of Bolt
reviewed literature. Reference
Reference
1.1. Carbon steel
Carbon steel ISO
ISO898-1
898-1 [86][86]
Serial Number Material of Bolt Reference
2. FullyFully
pearlitic high-carbon steel Lee etetal. [87]
1. 2. pearlitic
Carbon high-carbon
steel steel ISOLee
898-1al.[86][87]
3.3. SAE 10B38
SAE
steel and
10B38 steel
SAE 10B21
and SAE 10B21
steel Kodur et al. [88]
2. Fully pearlitic high-carbon steel steel Kodur
Lee et [87]
et al. al. [88]
4. Stainless steel, A2–70 and A4–70 Hu et al. [89]
3. 4. SAE 10B38 steel steel,
Stainless and SAE 10B21
A2–70 steel
and A4–70 KodurHu et
et al. [89]
[88]
5. Grade 10.9 high-performance (HP) steel Ban et al. [90]
4. 5. Stainless steel,
Grade 10.9 A2–70 and A4–70
high-performance (HP) steel Hu et et
Ban al.al.[89]
[90]
6. ASTM A325 bolt and A572 Grade 50 steel Cai et al. [91]
5. 6. GradeASTM
10.9 high-performance
A325 bolt A572 Gradesteel
and(Ti6Al4V)(HP) 50 steel Ban et al.
Caietetal. [90]
al.[92]
[91]
7. Titanium alloy Li
6. ASTM A325 bolt and A572 Grade 50 steel Cai et al. [91]
8.7. Titanium alloy
Aluminium (Ti6Al4V)
6082-T6 Li etet
Adeoti al.al.[92][93]
7. Titanium alloy (Ti6Al4V) Li et al. [92]
8. Aluminium 6082-T6 Adeoti et al. [93]
8. Aluminium 6082-T6 Adeoti et al. [93]
5.2. Nonconformity of the Material

5.2. Cui et al. [94]

5.2. Nonconformity
Nonconformity of inspected
of the fracture of a Ti-6Al-4V bolt that was generated during
the Material
the Material
the rolling
Cui et process of the threads (see Figure 12).
Cui et al.
al. [94]
[94] inspected
inspectedthethefracture
fractureofofa Ti-6Al-4V bolt
a Ti-6Al-4V that
bolt waswas
that generated during
generated the
during
rolling process of the threads (see Figure 12).
the rolling process of the threads (see Figure 12).

Figure 12. Crack generated on the threaded portion of a special screw upon rolling of the thread.
(a) Macro crack in the side view. (b) Macro crack in cross-section [94].
 Figure 12. Crack generated on the threaded portion of a special screw upon rolling of the thread. (a
Macro crack in the side view. (b) Macro crack in cross-section [94].

Machines 2023, 11, 212 During inspections, they found an excessive amount of oxygen, which 17 ofis
41 toxic fo
titanium alloy, in the vicinity of the fractured regions. This oxygen prevented the conver
sion from the alpha to beta phase by increasing the beta transition temperature and hin
deredDuring inspections, they
the crystallization byfound an excessive
increasing amount of oxygen,
the crystallization which of
temperature is toxic for There
the alloy.
titanium alloy, in the vicinity of the fractured regions. This oxygen prevented
fore, a coarse alpha phase was produced before the thread-rolling process, resulting i the conversion
from
poorthe alpha toand
ductility betaresistance
phase by increasing
to cracking the of beta transition
the temperature
alloy [94]. For such and hinderedalloy,
a sensitive the prope
crystallization by increasing the crystallization temperature of the alloy. Therefore, a coarse
shielding should be provided during the heat treatment processes; otherwise, atmos
alpha phase was produced before the thread-rolling process, resulting in poor ductility
pheric oxygen can accumulate in the alloy. Similarly, Jha et al. [95] analysed a case of shea
and resistance to cracking of the alloy [94]. For such a sensitive alloy, proper shielding
failurebe
should that occurred
provided at the
during thehead–shank
heat treatment transition
processes; ofotherwise,
a titaniumatmospheric
alloy fasteneroxygenused to joi
the flight spin motor to the subsystem. Scanning Electron Microscope
can accumulate in the alloy. Similarly, Jha et al. [95] analysed a case of shear failure that (SEM) analysis in
dicated at
occurred a the
crack at the thread
head–shank rootof
transition which supposedly
a titanium propagated
alloy fastener inwards
used to join and
the flight resulted i
spin
motor to the subsystem.
the separation Scanning
of the bolt head Electron
from theMicroscope (SEM) analysis
shank (full-length indicated
thread). The bolta crack
head is rec
at the thread root
ommended to bewhich supposedly propagated
manufactured using forging inwards
in theandregion
resultedofincoexistence
the separation of the tw
of
phases, which is below 995 °C for the Ti-6Al-4V alloy. However, the OpticaltoMicroscop
the bolt head from the shank (full-length thread). The bolt head is recommended be
manufactured using forging in the region of coexistence of the two phases, which is below
(OM) analysis showed that only the alpha phase existed in the vicinity of the fractured
995 ◦ C for the Ti-6Al-4V alloy. However, the Optical Microscope (OM) analysis showed
zone, which is indicative of the fact that processing was done above the recommended
that only the alpha phase existed in the vicinity of the fractured zone, which is indicative
temperature.
of Moreover,was
the fact that processing thedone
martensitic
above the structure is not observed
recommended temperature. in the shank the
Moreover, area, whic
martensitic structure is not observed in the shank area, which indicates that operation of Further
indicates that operation of martensitic formation was not conducted properly.
more, microscopic
martensitic formation analysis of the flow
was not conducted lines atFurthermore,
properly. the head-shank transition
microscopic region
analysis of reveal
theflow
the presence
lines atofthe
a sharp notch,transition
head-shank shown inregion Figurereveals
13, which promotes
the presence of fracture. In summary
a sharp notch,
shown in Figure of
a combination 13,improper
which promotes
material fracture. In summary,
processing, both ina terms
combination of improper and ge
of microstructure
material processing, both in
ometry, led this bolt to failure. terms of microstructure and geometry, led this bolt to failure.

Figure13.
Figure 13.Optical
Opticalmicrograph
micrograph showing
showing material
material flow
flow at the
at the junction
junction of head
of head and and shank
shank in the bolt
in the
[95].
bolts [95].

Moreover,
Moreover, Gong
Gonget al.
et [96] investigated
al. [96] the fracture
investigated of TP321
the fracture of stainless-steel anchor ancho
TP321 stainless-steel
bolts in seawater booster pumps used in nuclear power plants. The corrosion,
bolts in seawater booster pumps used in nuclear power plants. The corrosion, visible on thevisible o
fracture surfaces of the bolts, was promoted by the presence of chlorine (visible in energy
the fracture surfaces of the bolts, was promoted by the presence of chlorine (visible i
dispersive spectrometry (EDS) analysis). This chlorine was thought to be introduced into
energy dispersive spectrometry (EDS) analysis). This chlorine was thought to be intro
the material during the maintenance of the pump where some seawater could have been
duced into
sprinkled the
onto material
the pumps. during the maintenance
This resulted in the failure of
of the pump
the bolts whereSSC,
through some as seawater
visible could
have
by the been sprinkled
morphology onto the
of cracks pumps.
in the This resulted
axial plane in shown
of the bolt, the failure of the14,bolts
in Figure through SSC
obtained
as visible
through by theelectron
scanning morphology
microscopyof cracks
(SEM).in the axial plane of the bolt, shown in Figure 14
obtained through scanning electron microscopy (SEM).
Machines2023,
Machines 11,x212
2023,11, FOR PEER REVIEW 18of
18 of42
41

Figure 14. Microscopic images of cracks on the axial plane of the broken bolt: (a) branch cracks on the
right of14.
Figure theMicroscopic
axial plane,images
(b) cracks presentoninthe
of cracks a mid-axial plane
axial plane andbroken
of the (c) magnified
bolt: (a)branch
branchcracks
cracks[96].
on
the right of the axial plane, (b) cracks present in a mid-axial plane and (c) magnified branch cracks
5.3. Material Texture
[96].
In the selection of threaded fasteners for a specific application, the texture of the
fastener materials
5.3. Material Textureplays a significant role. The texture is affected by the different processes
performed on the fasteners during manufacturing, such as annealing, tempering, cold
In the selection of threaded fasteners for a specific application, the texture of the fas-
working, etc. Grimmer et al. [97], who analysed the behaviour of different-sized threaded
tener materials plays a significant role. The texture is affected by the different processes
fasteners, highlighted a difference in the engineering stress–strain curve between smaller
performed on the fasteners during manufacturing, such as annealing, tempering, cold
(S02 (NAS1351N00-4), S00(NAS1352NO2-6)) and larger fasteners (S04(NAS1352N04-8),
working, etc. Grimmer et al. [97], who analysed the behaviour of different-sized threaded
S06(NAS1352N06-10), S4(NAS1352N4-24)) which was probably due to differences in the
fasteners, highlighted a difference in the engineering stress–strain curve between smaller
grain orientations. The technique used to determine the grain orientation was the EBSD
(S02 (NAS1351N00-4), S00(NAS1352NO2-6)) and larger fasteners (S04(NAS1352N04-8),
mapping (Inverse Pole Figures). The grains are elongated with greater orientation in
S06(NAS1352N06-10), S4(NAS1352N4-24)) which was probably due to differences in the
the <111> direction, indicating the cold working performed on the fastener, while it is
grain orientations. The technique used to determine the grain orientation was the EBSD
equally distributed in the S02 fastener, indicating that, in the latter case, annealing was
mapping
performed, (Inverse
whichPole Figures).
resulted The grains set
in a distributed areof elongated with greater
grain orientations. orientation
This in the
greater number
<111>
of slipdirection,
directionsindicating
allowed the coldatworking
failure higher performed on thewhich
levels of strain, fastener,
is inwhile it is equally
accordance with
distributed
the engineering stress–engineering strain curve [97]. Similarly, Qi et al. [98] performed,
in the S02 fastener, indicating that, in the latter case, annealing was studied the
which resulted inofa Ti-6Al-4V
manufacturing distributedfasteners
set of grain
usedorientations.
in aerospace This
and greater number
different of slip direc-
applications due
tions allowed failure at higher levels of strain, which is in accordance
to their light weight and high strength. By paying close attention to the microstructurewith the engineering
stress–engineering
and grain orientation strain curve
during [97]. Similarly,
different processes Qiperformed
et al. [98] studied the manufacturing
on the alloy, they were ableof to
Ti-6Al-4V fastenerswith
produce fasteners used in aerospace
a yield strengthand different
higher applications
than that due to
of those used in their light weight
the literature. The

and highofstrength.
analysis texture By paying
after rollingclose attentionshowed
and drawing to the microstructure
a deformed alpha and grain
phase orientation
with 0110
during different processes performed on the alloy, they were able to produce fasteners


and 1 2 1 0 directions, as shown in Figure 15a. Then performing heading destroyed the
with a yield fibre
continuous strength higher
texture, andthan that of
solution those used
treating in the literature.
and ageing The analysis
(STA) transformed of texture
the grains into
after rolling
bimodal and drawing showed
microstructures, as shown a deformed
in Figure alpha phase with
15b. Hence, 0 1 1 0 isand
if the texture not 1considered
2 1 0 di-
rections,
during theas design
shown phase
in Figure 15a. Then
of bolts, performing
noticeable heading
differences can destroyed
be observed theincontinuous fibre
the stress–strain
texture,
behaviourandofsolution
the partstreating
[98]. and ageing (STA) transformed the grains into bimodal mi-
crostructures, as shown in Figure 15b. Hence, if the texture is not considered during the
design phase of bolts, noticeable differences can be observed in the stress–strain behaviour
of the parts [98].
Machines 2023, 11, Machines
212 2023, 11, x FOR PEER REVIEW 19 of 41

Figure 15.
Figure 15. SEM images SEM
after (a) images
drawingafter
and(a) drawing
rolling, andand rolling,
(b) after STAand
and(b)Inverse
after STA
poleand Inverse
figures afterpole figure
(c) drawing and rolling, and (d) heading and STA processes [98].
(c) drawing and rolling, and (d) heading and STA processes [98].

5.4. Viscoplasticity
5.4. Viscoplasticity
ViscoplasticityViscoplasticity
of materials shall of materials
be regardedshallasbeanregarded
additional as issue
an additional
when dealingissue when d
with threaded with threaded
fasteners, as it fasteners,
can lead toaslossit can lead to loss
of preload. of preload.
Different Different
materials havematerials
different have dif
viscoplasticGerman
viscoplastic behaviours: behaviours: German
guideline VDIguideline
2230 mentionsVDI 2230that,mentions
even in the that, even in the ab
absence
of external loads on the joint,
of external loads there canjoint,
on the be preload
there can lossbedue to embedment,
preload loss due tocreep or stresscreep or
embedment,
relaxation. Steel and aluminium
relaxation. Steel and display different
aluminium material
display responses,
different materiale.g.,responses,
embedding is embedd
e.g.,
more pronounced moreinpronounced
aluminiumin asaluminium
compared as to compared
steel. Furthermore, due to a difference
to steel. Furthermore, due to a differe
in the thermalthe expansion
thermal coefficient, the responses
expansion coefficient, of different
the responses ofmaterials vary [59].vary
different materials Den[59]. Den
Otter and Maljaars [12] studied the application of stainless-steel bolts on aluminium
and Maljaars [12] studied the application of stainless-steel bolts on aluminium p plates.
They observedThey the effects of viscoplasticity
observed the effects ofand temperatureand
viscoplasticity change on the preload
temperature changeloss on of
the preloa
the mentionedofbolts. Their investigation
the mentioned bolts. Theirshowed that there
investigation had been
showed thata there
significant loss aofsignifican
had been
preload as compared to the
of preload as case of carbon
compared steel
to the joint
case due to steel
of carbon the embedment
joint due tocreep, creep of creep,
the embedment
aluminium plate and the stress relaxation in stainless-steel bolts. Moreover, the
of aluminium plate and the stress relaxation in stainless-steel bolts. Moreover, the temperature
change studiesperature
showed that a temperature
change studies showeddrop of 30 a◦ Ctemperature
that resulted in adroppreload
of 30 loss
°Cofresulted
5% in in a pr
the joint. In addition, the testing of bolts with vibrational excitation showed a significant
loss of 5% in the joint. In addition, the testing of bolts with vibrational excitation sh
amount of preload loss dueamount
a significant to lateralofcontraction.
preload lossTherefore, a higher
due to lateral safety factor
contraction. (partiala higher
Therefore,
coefficient) must be used
factor to account
(partial for such
coefficient) mustpronounced
be used to preload
accountlosses [12].pronounced preload
for such
[12].
5.5. Corrosion
Corrosion5.5.
is another
Corrosionmaterial-dependent phenomenon that might induce failure in
threaded fasteners. Radouani et al. [99] performed numerical simulations on the corrosion
Corrosion is another material-dependent phenomenon that might induce fail
effects of bolted joints comprising low-alloy steel bolts and carbon-alloy steel plates by
threaded fasteners. Radouani et al. [99] performed numerical simulations on the corr
parameters from an electrochemical bath of 1 M HCl solution. They showed that the
effects of bolted joints comprising low-alloy steel bolts and carbon-alloy steel pla
corrosion rate is directly proportional to the size of the bolt and depends on the steel alloy.
parameters from an electrochemical bath of 1 M HCl solution. They showed that th
Moreover, for bolts which are in contact with water, such as those applied to turbines,
rosion rate is directly proportional to the size of the bolt and depends on the steel
a highly corrosion-resistant material must be selected [99]. Casanova and Mantilla [45]
Moreover,
report the failure of boltsfor bolts which
connecting are in to
a turbine contact withofwater,
the shaft such as those
the alternator. Theyapplied
found to turbi
highly corrosion-resistant material must be selected [99]. Casanova
corrosion at the roots of some threads and regarded this occurrence as a possible reason and Mantilla [4
for the failure they documented. As a matter of fact, corrosion enables fatigue cracks to They
port the failure of bolts connecting a turbine to the shaft of the alternator.
corrosion at thelow
grow even at comparatively roots of some
levels threads
of stress. A and regarded
possible this to
solution occurrence
the problemas a ispossible r
for the failure
to change the material of the they
boltsdocumented. As atomatter
to stainless steel of fact,
minimize corrosion
corrosion [45].enables fatigue cra
Similarly,
Zhang et al. [86] studied the risk of galvanic corrosion in ASTM A325 bolts (a material problem
grow even at comparatively low levels of stress. A possible solution to the
whose mechanicalchange the materialare
characteristics of comparable
the bolts to tostainless
that of steel to minimize
8.8 class corrosionto[45]. Sim
screws according
Zhang et al. [86] studied the risk of galvanic corrosion in ASTM A325 bolts (a ma
Machines 2023, 11, 212 20 of 41

[ISO 898-1] when coupled with plates made of a martensitic stainless-steel ASTM A1010.
They found out that, in the case of heavy saline exposure, the galvanic corrosion rate of
A325 bolts was significantly higher when used in combination with A1010 steel than when
used with an A588 weathering steel. The study suggests that a reduction of the galvanic
corrosion rate can be achieved by painting the plates at the sites of application of the bolts,
thereby reducing the cathode-to-anode area ratio [100]. Furthermore, the lubricant type
also affects the corrosion characteristics of bolted joints. In fact, aluminium connections can
undergo corrosion due to some lubes, which limits the number of suitable lubricants [12].

5.6. Supplier
Another factor leading to bolt failure is the inadequate quality of bolts. Abid et al. [101]
applied the same torque to M12 bolts from different suppliers. The torques required to
achieve the required preload were the same for grade 8.8 and grade 10.9 bolts, whereas
another stud from an unknown supplier yielded at a lower value of torque. The reason
was attributed to the overload because of the unknown property class of that stud. Some-
times, the combination of overloading with other factors leads to the failure of bolted
connections [101]. Lacalle et al. [102] reported the failure of a bolt in a scaffolding. The
failure analysis of the bolt showed that the crack was generated inside the bolt due to some
environmental factors, such as corrosion. After the crack generation, propagation was
carried out by the overloading on the bolt, which eventually resulted in the failure of the
bolt during the dismantling procedure.

6. Fretting Fatigue
Fretting fatigue is another phenomenon which can result in the failure of threaded
fasteners. Fretting fatigue is generated in the presence of moving frictional contacts which
promote the formation of micro cracks that, in turn, lead to macro cracking at stress
concentration regions [103]. Such an issue has been a challenge for threaded fasteners in
the past and, as of today, it is still a point of concern. Walker [104] reported that failures
of aircraft occur through fretting at the points which are connected through bolts, pins
and rivets. Among these, bolted joints are sites of particular concern. Sandifier [105]
studied the fretting fatigue of aluminium lap joints and suggested design improvements to
improve their fretting resistance. According to the assessment carried out by the authors, a
material replacement was not feasible due to other design constraints. The most favourable
approach was to deploy compressive residual stresses on the joint surface by shot peening
to prevent crack growth, along with adhesive application on the contacting surfaces before
the assembly of the joint to prevent fretting initiation [105]. Wagle and Kato [106] studied
the fatigue behaviour of bolted joints used in combination with 2024-T3 aluminium plates.
They studied the effect of increasing the initial torque from 0 (pin loading) to 8 Nm. They
found that with increased applied torque, the fatigue life of the bolted joint increased up to
a certain level. However, at much higher levels of tightening torque, the fretting fatigue
started affecting the joints in the vicinity of the bolt hole. The authors also used ultrasonic
waves to detect the fretting crack in the assembled position at 1.8 to 2.1 mm away from
the bolt hole [106]. Chakherlou et al. [107] also studied the double-shear lap joint made by
2024-T3 aluminium alloy under fatigue conditions. The initial torques applied were 0.25, 2
and 4 Nm, as shown in the S–N curve reported in Figure 16. They reported that increasing
the amount of applied torque led to an increase in fatigue life. However, below a certain
level of external loading, failure occurs due to fretting fatigue. As a matter of fact, at about
10 kN of external loading, failure was due to fretting fatigue, whereas at 18 kN fretting
was not observed. The reason is that load is transferred by the friction of the plates at a
low level of excitation, preventing the bolt shank from coming into contact with the hole
surface. Nonetheless, at elevated levels of excitation, both friction between the plates and
the bolt shank transfer the load. Therefore, based on the outcomes of this study, attention
must be paid to the fretting phenomenon, especially at low levels of excitation.
 both friction between the plates and the bolt shank transfer the load. Therefore, ba
Machines 2023, 11, 212 the outcomes of this study, attention must be paid to the fretting phenomenon,
21 of 41 esp
at low levels of excitation.

Figure 16. S-N curve for the double shear lap joint at different torque levels (4 N·m, 2 N·m and
0.25 N·m)
Figure 16.without lubrication
S-N curve [107].
for the double shear lap joint at different torque levels (4 N·m, 2 N·m
N·m) without lubrication [107].
Significant efforts have been put into studying the fretting fatigue issue since the
beginning of the 20th century [108]: as early as 1964, Ungar [109] used experimental
Significant
techniques to find efforts
the rate have
of decaybeen put into
of joints’ studying
structural the fretting
responses to study fatigue
the dampingissue since
ginning of the
characteristics 20th century
of riveted [108]:
joints, which as early
could as 1964,
be increased Ungar [109]
by loosening used
of rivets. Heexperiment
also
found out that, upon loosening the rivets, fretting fatigue was induced
niques to find the rate of decay of joints’ structural responses to study the dampin on the joint. Rogers
and Boothroyd used an experimental technique to produce the loops of fretting response
acteristics of riveted joints, which could be increased by loosening of rivets. He also
(friction force vs displacement amplitude) for metal interfaces [110].
out that, upon loosening
Eriten et al. [111] developedthea new
rivets, fretting
apparatus fatiguethe
to measure was induced on
fretting-fatigue the joint. Rog
behaviour
Boothroyd usedby
of bolted lap joints ancompensating
experimental technique
for the misalignment to produce the loops of operation
and the high-frequency fretting respon
with respect to previous setups [112–114]. The misalignment
tion force vs displacement amplitude) for metal interfaces [110]. accuracy had been improved
by using a high-accuracy internal sensor in the piezo translator (PZT) and a 3-axis load
Eriten et al. [111] developed a new apparatus to measure the fretting-fatigue
cell for force measurements. Moreover, the design of the apparatus allows service at high
iour of bolted
frequencies, lap joints byamplifier
so a high-frequency compensating
can be used forwith
thePZTmisalignment and theuphigh-fre
to perform calculations
operation
to 3–4 kHz. with
Using respect to previous
this apparatus, setups
the fretting [112–114].
behaviour The misalignment
of single-bolt aluminium and accuracy
steel ha
improved by using
joints was studied. Theastiffness
high-accuracy
and energy internal
dissipation sensor
from in thethe piezofretting
obtained translator
loops(PZT) a
was in
axis agreement
load cell forwith the measurements.
force existing literature [115,116].
Moreover, Thetheenergy dissipation
design of thewas more
apparatus allo
pronounced in aluminium joints compared to steel. Moreover, the shakedown process was
vice at high frequencies, so a high-frequency amplifier can be used with PZT to p
greater in aluminium with respect to the steel joint.
calculations
Mutoh and upJayaprakash
to 3–4 kHz. Usinganalysed
[117,118] this apparatus,
the frettingthe fretting
failure behaviour
of various of single-
steel alloys
uminium
by means ofand steel joints
a tangential stresswas studied. Thestress
range–compressive stiffness
rangeand energy
(TSR-CSR) dissipation
diagram. They from
developed a generalized TSR-CSR diagram using the normalized
tained fretting loops was in agreement with the existing literature [115,116]. The tensile and compressive
stress ranges for various steel alloys, as shown in Figure 17. The meaning of the diagram
dissipation was more pronounced in aluminium joints compared to steel. Moreo
can be explained as follows: each point is obtained by FE calculation of the compressive
shakedown
and tangentialprocess
normalised wasstress
greater
ranges;in then,
aluminium with
if the point fallsrespect
above the to experimentally
the steel joint.
Mutohfretting
determined and Jayaprakash [117,118]
fatigue limit line, the jointanalysed
is subject tothe fretting
fretting failure
fatigue of various
failure. If the stee
point
by falls below
means the line, the stress
of a tangential joint is range–compressive
safe. They assessed the stress diagram on several
range types ofdiagram
(TSR-CSR)
contact geometries, including bolted joints. This generalized
developed a generalized TSR-CSR diagram using the normalized tensile diagram could then be used
and comp
along with FEA to predict the behaviour of bolted joints under fretting fatigue conditions,
stress ranges for various steel alloys, as shown in Figure 17. The meaning of the d
limited to the case of steel components.
can be explained as follows: each point is obtained by FE calculation of the comp
and tangential normalised stress ranges; then, if the point falls above the experim
determined fretting fatigue limit line, the joint is subject to fretting fatigue failure
point falls below the line, the joint is safe. They assessed the diagram on several t
contact geometries, including bolted joints. This generalized diagram could then b
along with FEA to predict the behaviour of bolted joints under fretting fatigue con
limited to the case of steel components.
 Machines2023,
Machines 11,x212
2023,11, FOR PEER REVIEW 22 of 22
41 of 42
Machines 2023, 11, x FOR PEER REVIEW 22 of 42

Figure 17. Generalized TSR-CSR diagram for various steels [117,118].

Figure 17.Generalized
Figure17. GeneralizedTSR-CSR
TSR-CSRdiagram forfor
diagram various steels
various [117,118].
steels [117,118].
6.1.Experimental
6.1. ExperimentalAnalyses
Analyses
6.1. Experimental Analyses
Juoksukangasetetal.
Juoksukangas al.[119]
[119]developed
developedan anexperimental
experimentalapparatus
apparatusand andmodified
modifiedititfor for
Juoksukangas
measuring
measuring thefretting
the et fatigue
fretting al. [119]ofof
fatigue developed
boltedjoints,
bolted an as
joints, experimental
asshown
shownininFigure apparatus
Figure 18.The
18. and
The flatmodified
flat specimens
specimens it for
measuring
to
tobe
beconnected the fretting
connected by boltsfatigue
bybolts are
areheld
heldofatat
bolted
one
oneend joints,
end by as
bythe shown
themain in Figurewhereas
mainclamping,
clamping, 18. Thevibrational
whereas flat specimens
vibrational
to be
loadingconnected by bolts are held at one end by the main
loading is applied at the free end, by means of an eccentric mechanism. At 30 mmaway
is applied at the free end, by means of an eccentric clamping,
mechanism. whereas
At 30 mm vibrational
away
loading
from
fromthetheis
mainapplied
main at theaabolted
clamping,
clamping, free end,
bolted jointbyisismeans
joint provided.of an
provided. eccentric
Strain
Strain gauges
gauges mechanism.
are
areconnected
connected Atto30 mm
tothe
the away
bolt
bolt
from
as
aswellthe
well asasmain
totothe clamping,
theupper
upper a bolted
specimen
specimen to tojoint
monitor is provided.
monitor thethe
stresses Strain
stressesandand gauges
the forces
the are connected
acting
forces on the
acting to the
sys-bolt
on system.
the
The
as
tem.displacements
well The are calculated
as displacements
to the upper specimen using theusing
to monitor
are calculated Digital
theImage
the stresses
DigitalCorrelation
and the
Image (DIC)
forcestechnique.
acting
Correlation (DIC) The
on tech-
the sys-
results
nique.
tem. showed
The thatshowed
Thedisplacements
results fretting decreased
that
are frettingthe
calculated fatiguethe
decreased
using life when
fatigueincreasing
theDigital life when
Image the preload(DIC)
increasing
Correlation or the
the pre-
tech-
bulk
nique.stresses.
load orThetheresults
bulk stresses.
showed that fretting decreased the fatigue life when increasing the pre-
load or the bulk stresses.

Figure 18. The experimental apparatus used for measuring fretting fatigue in bolted joints [119].
Figure 18. The experimental apparatus used for measuring fretting fatigue in bolted joints [119].

Jiménez-Peña
Figure
Jiménez-Peña etetal.
al.[120]
18. The experimental [120] studied
apparatus
studied thefor
used
the fretting-fatigue response
measuring fretting
fretting-fatigue ofofSingle
fatigue
response inSingle
boltedLap Bolted
joints
Lap [119].
Bolted
Joints(SLBJ)
Joints (SLBJ)made
madeofofS500MC
S500MCsteel
steel(a(alow-alloy,
low-alloy,high-strength
high-strengthsteel
steelwith
withUTS
UTS≈≈650
650MPa).
MPa).
They applied
They Jiménez-Peñatension–tension fatigue
et al. [120]fatigue
applied tension–tension studied cycles on the joint by
the fretting-fatigue
cycles means
response
on the joint by means of a universal
of a of testing
Single Lap
universal Bolted
testing
machine.
Joints They
(SLBJ) madealsoof used
S500MCFEAsteel
modelling to further
(a low-alloy, study thesteel
high-strength phenomenon.
with UTS ≈Fretting
650 MPa).
They applied tension–tension fatigue cycles on the joint by means of a universal testing
machine. They also used FEA modelling to further study the phenomenon. Fretting
Machines 2023, 11, 212 23 of 41

machine. They also used FEA modelling to further study the phenomenon. Fretting
fatigue turned out to be the dominating failure mode and it decreased with increasing the
initial preload. The preload force tended to affect the size of the sticking region and the
position of the stress concentration. Moreover, the authors applied different multiaxial
criteria (SWT, Fatemi-Socie and others) to forecast the site of initiation of the fretting-fatigue
crack: all of them agreed upon the same location of the initiation site, i.e., around the
centreline of the specimen, at a certain distance from the hole edge [120]. Li et al. [121] used
a piezoelectric actuator to provide controlled displacement, as explained in a previously
analysed paper [111]. Nonetheless, they improved the apparatus by using one laser beam
along with a specific combination of the prism. By using this, they were able to measure the
relative displacements more accurately and at a comparatively low cost (some studies used
two laser beams). Moreover, using a spring leaf, the motion of the moving contact surface
was guided, hence giving more control over the motion. They measured the effect of bolt
preload and excitation amplitude. The fretting hysteresis loops changed significantly in
the case of changing bolt preload. An increase in bolt preload resulted in an increase in
friction force and a decrease in sliding amplitude. The fretting hysteresis loops changed
from elliptical to parallelogram by increasing the excitation amplitude [121]. Szlosarek and
Kröger [122] found an interesting behaviour while evaluating the fatigue performance of
M22 screws combined with 5 mm-thick steel sheets (S355MC). They found out that at lower
preload levels (0–30 kN), failure of the bolted connection took place at the bolt hole, while
for higher preloads (110 and 225 kN), the crack initiates at the pressure zone, indicating the
presence of the fretting phenomenon.

6.2. Numerical Analyses

Over the years, many authors put forth procedures for estimating the fretting fatigue
behaviour of bolted joints which make use of numerical methods. Ferjaoui et al. [123]
used FEA coupled with continuum damage mechanics to estimate the number of cycles
to crack initiation in double-lap bolted joints (DLBJ). Figure 19a shows the FEA setup
used by Ferjaoui et al., along with the relevant boundary conditions. In the first load step,
contact between the pairs was established where the translational degrees of freedom were
restrained in Al plates’ left and right-hand sides. In the second load step, the axial stress
was applied to the right-hand side by removing the constraints. The mesh element was
8-node linear (C3D8R) with finer mesh at the contacts of 0.1 mm, as shown in Figure 19b.
The estimated crack initiation life is between 80% and 90% of the total experimental life
for the partial slip condition, showing a good agreement between the combined analytical-
numerical model and the experimental data [123]. Similarly, many researchers have tried
to tackle this problem using different approaches, including finite element methods (FEM),
boundary element methods (BEM), finite-discrete element methods (FDEM) and molecular
dynamics (MD) [124–127]. Armand et al. [128] considered the effect of surface roughness
on the fretting behaviour of assembled structures using a BEM-based contact solver. They
used two different time scales and space scales. This procedure enabled them to decouple
the contact and the dynamics model, which in turn made it possible to study surface
roughness more closely. Li and Yang [129] also performed a numerical analysis of the
bolted joints wherein they introduced the k parameter, which represents the ratio between
the bolt pretension and the shear load, and used this parameter to investigate the fretting
behaviour of the bolted joint. They determined experimentally that when k is unity, the
fatigue behaviour of the joint is dominated by the stress concentration. On the contrary,
when k increases, fretting becomes the dominating factor.
As a final remark, due to the complex nature of the fretting-fatigue phenomenon, it is
paramount to leverage all the available design tools (FE, combined with experimentation
and analytical calculations based on well-established multiaxial criteria) in order to correctly
define the parameters of the joint, based on the specific boundary conditions.
Machines2023,
Machines 11,x212
2023,11, FOR PEER REVIEW 24ofof42
24 41

Figure
Figure 19.
19. FEA
FEAmodel
modelshowing
showing(a)
(a)the
theboundary
boundaryconditions
conditionsof
ofDLBJ,
DLBJ,(b)
(b)mesh
meshgeneration
generationin
inthe
the
geometry [123].
geometry [123].

As a final
7. Fatigue remark, due to the complex nature of the fretting-fatigue phenomenon, it
Failure
is paramount
It is widely leverage
to all the
known that available
fatigue failuredesign tools
may take (FE,when
place combined with parts
structural experimenta-
undergo
tion and analytical calculations based on well-established multiaxial criteria)
cyclic loading [130]. Bolted connections are subjected to variable loadings that can in order to
be cyclic
correctly
in naturedefine the parameters
in automobiles, of the joint,
aeroplanes based
as well on the specific
as machinery boundary
in general. The conditions.
mechanism of
fatigue failure is usually regarded as a three-stage process. At first, the crack initiates in the
7.material
Fatiguedue Failure
to the presence of defects or at stress concentration sites, under the action of
It is widely
repeated loadingknown that fatigue
and unloading failure may
sequences. The take
second place
stagewhen structural
involves parts undergo
the propagation of a
dominant
cyclic loading crack. TheBolted
[130]. final stage consistsare
connections of the full rupture
subjected of theloadings
to variable part duethat to the
canfactbe that
cy-
the in
clic remaining
nature incross-section
automobiles,isaeroplanes
no longer able to withstand
as well as machinery the design loads.The
in general. Themechanism
presence of
thefatigue
of crack can be due
failure to different
is usually reasons,
regarded as aeach pointingprocess.
three-stage to a different mechanism
At first, the crack[131–133].
initiates
The reasons for the formation of a crack in fatigue failure
in the material due to the presence of defects or at stress concentration sites, under thecan be due to the geometry of
bolted connections, the profile of the loading on the joint,
action of repeated loading and unloading sequences. The second stage involves the prop- material defects and less than
requiredofbolt
agation preload, crack.
a dominant etc. Extensive
The finalresearch has been
stage consists performed
of the to find
full rupture of the relationship
part due to
between
the fact that thesetheparameters
remaining and the fatigue
cross-section is life
no of the specimen.
longer able to withstand the design loads.
As mentioned
The presence of theincrack
the discussion
can be due of to
improper
different design
reasons,parameters, the application
each pointing to a differentof an
appropriate preload through the initial torque is mandatory
mechanism [131–133]. The reasons for the formation of a crack in fatigue failure can be for the proper operation of
bolted joints. It is well-known that the application of insufficient
due to the geometry of bolted connections, the profile of the loading on the joint, material preload results in reduced
fatigueand
defects life less
due thanto the comparatively
required higheretc.
bolt preload, stresses
Extensiveseenresearch
by the screw
has beenfor given
performedexternal
to
find the relationship between these parameters and the fatigue life of the specimen. the
axial loads [48]. In some cases, as documented above, the preload decreases during
loading cycles due in
As mentioned tothe
thediscussion
transverseof contraction
improper of the plates,
design whichthe
parameters, affects the contact
application of anof
the bolt and plates [134]: this occurrence is particularly relevant
appropriate preload through the initial torque is mandatory for the proper operation of in the case of steel bolts
applied
bolted to aluminium
joints. plates. Based
It is well-known that the onapplication
the aforementioned
of insufficientconsiderations,
preload results the fatigue
in re-
duced fatigue life due to the comparatively higher stresses seen by the screw increasing
strength of the preloaded joint is better than that of the non-preloaded joint, and for given
the amount
external axialofloads
preloading
[48]. Inusually determines
some cases, an increase
as documented in the
above, thefatigue
preload lifedecreases
of the joint for
dur-
given boundary conditions [135,136]. Chakherlou et al. [137]
ing the loading cycles due to the transverse contraction of the plates, which affects the performed fatigue testing of
M5 × 0.8 hexagon head steel bolts which were preloaded at 0, 244, 3409 and 6818 N. In
contact of the bolt and plates [134]: this occurrence is particularly relevant in the case of
another study, Chakherlou et al. [107] performed testing on a bolted connection (hexagon
steel bolts applied to aluminium plates. Based on the aforementioned considerations, the
head M6 × 1, class 10.9 bolts) comprising aluminium plates at the initial tightening torques
fatigue strength of the preloaded joint is better than that of the non-preloaded joint, and
of 0.25, 2 and 4 Nm. In both cases, they found that the fatigue strength of the bolted connec-
increasing the amount of preloading usually determines an increase in the fatigue life of
tions increased with increasing the amount of initial preload [107]. Esmaeili et al. [138,139]
the joint for given boundary conditions [135,136]. Chakherlou et al. [137] performed fa-
used FEA to understand the effect of different amounts of initial torques on the bolts (1, 2.5
tigue testing of M5 × 0.8 hexagon head steel bolts which were preloaded at 0, 244, 3409
and 5 Nm) and verified the models experimentally. The results showed that the increase of
and 6818 N. In another study, Chakherlou et al. [107] performed testing on a bolted con-
tightening torque increases the fatigue strength of the tested aluminium double-lap bolted
nection (hexagon head M6 × 1, class 10.9 bolts) comprising aluminium plates at the initial
joints [138,139]. Abazadeh [140] performed numerical simulations on single-lap bolted
tightening torques of 0.25, 2 and 4 Nm. In both cases, they found that the fatigue strength
joints and validated them experimentally. He found out that at a remote stress level of
of the bolted connections increased with increasing the amount of initial preload [107].
20 MPa (cyclic load amplitude), the increase in tightening torque from 0 to 5 Nm increased
Esmaeili
the fatigue et al.
life[138,139] used
of the joint. InFEA to understand
the same study, Abzadeh the effect of different
showed that foramounts of initial
higher tightening
torques on the bolts (1, 2.5 and 5 Nm) and verified the models
torques of 5 to 8 Nm, the increase in tightening torque decreases the fatigue life of the experimentally. The results
joint.
showed
This decrease was attributed to the fretting behaviour of the joint, as confirmed by thetested
that the increase of tightening torque increases the fatigue strength of the crack
aluminium
being produced double-lap
away frombolted joints
the hole[138,139].
[140]. Abazadeh [140] performed numerical sim-
ulations on single-lap bolted joints and validated them experimentally. He found out that
Machines 2023, 11, 212 25 of 41

Furthermore, the influence of the geometry of the bolted connections on fatigue

life has also been investigated by many scholars. Munse et al. [141] investigated the
influence of changing the grip length of 1 in. high tensile bolts from 1 3/4 to 3 3/4 in.
(approx. 44 to 95 mm) and reported that there is no change in fatigue strength. The fatigue
strength obtained for all the specimens was 18,000 psi at 2,000,000 cycles. Griza et al. [142]
investigated the influence of bolt length in M24 × 3 bolt studs used in a gas compressor for
assembling flanges to housing. The experimentation was performed on nine short bolts
with a length of 120 mm and eight long bolts of 173 mm in length. A spread sleeve was
used with the longer bolt studs. The authors’ findings indicated that the increase in the bolt
length increases the fatigue life of the bolted connection. The increase in the length of the
bolts from 120 to 173 mm increased the fatigue limit at 5 million cycles by 8.6%. However,
this difference might be attributed to the introduction of spread sleeves in longer bolts,
which increases the member’s stiffness and decreases the bolt stiffness [142].
Moreover, the profile of load application for fatigue behaviour must also be considered.
The two parameters, namely stress range (difference between maximum and minimum
stresses) and the stress ratio (ratio of minimum to maximum stress) are of paramount im-
portance. For the stress range, it was shown by Chakherlou et al. [107] that the mechanism
of failure is different for the cases of low (fretting failure) and high (fatigue failure) load
ranges, as discussed above. As for the stress ratio, there exists an inverse relationship
between the stress ratio and fatigue strength. Birkemoe et al. [143] observed that there
is an increase in the value of fatigue life when the stress ratio is decreased. Similarly,
Atzori et al. [144] made a statistical analysis of aluminium bolted joints and observed that
a change in the stress ratio from 0 to −0.5 in bearing-type joints increased fatigue strength
by 50%. Zhou et al. [145] investigated composite bolted joints subjected to fatigue loading.
They developed a modified version of the S-N curve for their model, by considering the
stress ratio and the ply angle. They also reported that for constant ply angle, the slope of the
S-N curve increased by increasing the equivalent stress ratio. Therefore, also in composite
bolted joints, the fatigue life decreases by increasing the stress ratio [145].
Furthermore, materials processing including surface treatments can affect the fatigue
life of bolted connections. It has been studied that zinc coating through hot dip galvanizing
can have an adverse effect on the fatigue strength of the bearing type in which bolts are in
clearance holes [136]. However, if it is kept at a certain thickness, then the fatigue life is not
affected [146]. The generation of a hole can be done through different processes including
cold working and hot working. Eurocode 3 provides FAT classes for only the drilled holes
where each class has different fatigue curves. However, Cicero et al. [147] found that the
fatigue behaviour of bolted connections changes by altering the hole generation techniques.
They compared drilling with three thermal hole-generation techniques including oxyfuel
cutting, plasma and laser cutting. They reported that FAT90 can be used for both drilled as
well as oxyfuel-cut holes. On the contrary, for holes generated through laser and plasma
cutting, FAT63 must be used (applicable to <15 mm thickness) [147].
The use of threadlockers also affects the fatigue life of bolted connections. Drag-
oni [148] investigated the fatigue life of M10 steel studs (class 10.9) with and without
threadlocker. Loctite 243 was used as a threadlocker in this study. Fatigue testing was
performed on high (11 ± 10 kN) and low (8.8 ± 8 kN) levels of force. In both cases, the
fatigue life was improved by the use of the threadlocker by a ratio of mean life being 3.5 for
the high load range and 2 for the low load range [148].
Efforts have been made to study the fatigue behaviour of the bolted joints with respect
to the geometry and design of bolted joints. Taylor and Pan [149] performed a numerical
investigation to find out the stress and strain at the first thread of the joint by applying
fatigue loading. Using these stresses and strains, the multiaxial fatigue theory and critical
plane approach were employed to determine the failure location. For M14 metric threads
with a 0.29 mm root radius, the failure was found to originate at 18◦ from the centerline
of the thread root and inclined at 45◦ to the thread root surface. This methodology can be
employed to find the failure location of other thread sizes [149]. Similarly, O’Brien and
Machines 2023, 11, 212 26 of 41

Metcalfe [48] investigated the high-stress concentration regions in fasteners which lead to
fatigue failure of bolted joints. They inferred that work on reducing the stress concentration
regions has been performed in the past but is seldom applied in the mass production of
fasteners. They suggested using an elliptical shape for the underhead-shank transition to
reduce stress concentration. The thread should be roll-formed to achieve a good shape
of threads and to have residual compressive stresses. The nuts should be designed so as
to minimize stress concentration at the first engaged thread, and there should be threads
above and below the nut. Guide surfaces should be present with fitted fasteners with
generous radii. Hareyama et al. [150] devised a method for analysing the working loads on
the bolts by simulation of the actual machine environment. By this approach, the designer
is able to predict the strength and durability of bolted joints. Strain gauges were used to
monitor the strains on the bolts and analytical equations were proposed to convert these
strains into loads. A stress history diagram was obtained from the experimentation. This
diagram was then used to estimate the fatigue life of the connections.
Sorg et al. [151] proposed a local approach to predict the durability of bolted con-
nections subjected to fatigue loading. The approach based on normal forces could not
predict well the durability of the bolted connections subjected to fatigue loading. Hence,
Schneider’s method was improved to predict the multi-axial, time-dependent problems
of fatigue loading in screws. The presented method gives the values of damage in the
first load-bearing thread of the screw over the circumference of the thread [151]. Finally,
Nassar and Li [152] proposed another model for determining stresses in high-cycle fatigue.
Using such a model, the S-N curves were constructed with the help of experimental data of
ultra-high strength fasteners (grades 14.8, 15.8 and 16.8) and standard fasteners (grades
12.9 and 10.9). The normalized fatigue strength of UHS fasteners was then compared with
that of standard fasteners.

8. Overload
Bolted connections are designed according to the loading conditions that might occur
in the specific application. If the loads on the bolted connections exceed the limit, the
bolted connections fail, resulting in detachment of the plates, fracture of the bolts, or both.
In addition, if the initial torque of the bolts is increased beyond the design limitations,
the bolts can also fail upon assembly [153]. Overloading of the joint is driven by many
factors. One of the reasons is an incorrect application of lubricant on bolts. Eliaza et al. [154]
investigated the failure of bolts in a helicopter main drive plate assembly. Their analysis
showed that the incorrect application of Molykote lubricant spray, an anti-seize material,
was the reason for failure. Due to that reason, the same value of torque resulted in axial
force being 1.7 times higher than the required level. This was greater than the yield point of
the bolt. As a result of this investigation, it was suggested to apply the anti-seize product
only on the threads, thus explicitly excluding the underhead surface of the screw, in order
to keep the total friction coefficient of the joint within the desired range [154]. The bolt
failure analysis of Stevenson et al. [58] also showed that ductile overload was the reason
for failure. This was due to a deviation in the frictional characteristics of the joints.
In some circumstances, bolts may fail through simple static overloading. Xue et al. [155]
investigated the explosion of a reactor in which the bolts were seen fractured just after
the explosion. The investigation revealed that the bolts and the pressure vessel met all
the design parameters. Therefore, the only possibility of failure was overload in the pres-
sure vessel which resulted in tensile overload in the bolts and led to the failure of the
bolts and hence the reactor [155]. Moreover, Chen et al. [156] carried out an investigation
into the failure of rock bolts in a coal mine. The main outcome of their study consisted
of demonstrating that the failure of the majority of the analysed bolts was due to stress
corrosion cracking, whereas one of the bolts was fractured due to simple overloading due
to undesired bending stresses applied to the bolt [156]. Neidel et al. [157] reported that
some bolts failed due to torsional overloading during the final assembly stage of a gas
 pressure vessel which resulted in tensile overload in the bolts and led to the failure of the
bolts and hence the reactor [156]. Moreover, Chen et al. [157] carried out an investigation
into the failure of rock bolts in a coal mine. The main outcome of their study consisted of
demonstrating that the failure of the majority of the analysed bolts was due to stress cor-
Machines 2023, 11, 212 rosion cracking, whereas one of the bolts was fractured due to simple overloading27due to
of 41
undesired bending stresses applied to the bolt [157]. Neidel et al. [158] reported that some
bolts failed due to torsional overloading during the final assembly stage of a gas turbine
engine.engine.
turbine The reason for torsional
The reason overloading
for torsional was attributed
overloading to theto
was attributed non-conformity of the
the non-conformity
assembly procedure.
of the assembly procedure.

9.9.Corrosion
Corrosionand andOtherOtherEnvironmental
EnvironmentalEffects Effects
Stresscorrosion
Stress corrosion cracking
cracking (SCC)
(SCC) is regarded
is regarded as a frequent
as a frequent root of
root cause cause of in
failure failure
boltedin
boltedThe
joints. joints. The presence
presence of stresses,
of stresses, specific
specific bolt bolt materials
materials and a corrosive
and a corrosive environmentenvironment
are the
are thethat
factors factors that combine
combine to bring
to bring about about
this failurethismode,
failure mode,
which is which is brittle
brittle in naturein nature
and and
usually
usually
results in results in catastrophic
catastrophic failure of thefailure of the components,
components, as it usually as goesit undetected
usually goes undetected
prior to final
prior tosince
rupture finalitrupture
can takesince
placeit in
can
thetake place of
presence in mildly
the presence
corrosive of mildly corrosive
environments environ-
[158]. It is
aments
common [159].
typeIt of
is failure
a common type of failure
in underground rockinmines
underground
where therock mines whereconditions
aforementioned the afore-
mentioned
are conditions
easily fulfilled. are aeasily
In 2002, studyfulfilled.
showed In that2002,
mosta failures
study showedoccurringthatinmost failures
rock bolts were oc-
curring
due to SCC in rock
[159].bolts were
Crosky etdue to SCC
al. [160] [160]. Crosky
performed et al.
a failure [161] performed
analysis on 44 boltsawhich
failurefailed
analysis
in
four different underground coal mines in Australia. Most of the
on 44 bolts which failed in four different underground coal mines in Australia. Most of bolts that they investigated
were fractured
the bolts due investigated
that they to stress corrosion crackingdue
were fractured (SCC), as indicated
to stress corrosion bycracking
the discoloured
(SCC), as
region
indicatedin thebyfracture surfaces.region
the discoloured Moreover,
in thethe majority
fracture of the Moreover,
surfaces. bolts showed thebending,
majority from
of the
which it can bebending,
bolts showed estimated thatwhich
from bending
it can stresses were the
be estimated thatreason
bendingfor stresses
initiating SCCthe
were [160].
rea-
Similarly, Escobar SCC
son for initiating et al.[161].
[161] Similarly,
investigated the failure
Escobar of a submersible
et al. [162] investigatedpump system.
the failure of aThe
sub-
failure
mersible waspumpattributed to the
system. Thebolt which
failure was fractured
attributed duetotothe
SCC,boltaswhich
shownfractured
by the nucleation
due to SCC, at
the high-stress concentration and the presence of rust on the fracture
as shown by the nucleation at the high-stress concentration and the presence of rust on surfaces. The fracture
surface is shown
the fracture in Figure
surfaces. 20 withsurface
The fracture an arrow pointing
is shown to the 20
in Figure nucleation
with an arrowregion, and theto
pointing
propagation
the nucleation region shows
region, andthethepresence
propagation of corrosion.
region shows the presence of corrosion.

Figure20.
Figure 20. Fracture surface
surface at
at30×
30×showing
showing thethe
nucleation zone
nucleation indicated
zone by the
indicated by arrow and the
the arrow andprop-
the
agation zone
propagation showing
zone the the
showing red red
rustrust
[162].
[161].

Craig et al. [162] investigated the failure of several rock bolts coming from two hundred
mines located in Australia which were made from X-grade steel with a minor diameter of
21.7 mm. Of all the rock bolts, 70% had failed due to SCC, the rest of them failing from
pitting corrosion at local sites. Chen et al. [156] also investigated the failure of six HSAC
840 steel rock bolts from underground coal mines which had failed from the end of the
bolt in the threaded region. Failure of half of the bolts was initiated from the thread root,
while for the other half the failure was due to cracks initiated as a result of bolt bending.
Among the examined six bolts, five had failed due to SCC, as suggested by the presence of
Machines 2023, 11, 212 28 of 41

a 1.5–4.5 mm discoloured region at the origin of the fracture. The rock bolts usually used in
Australian rock mines (M24-3.0 × 2400) were given 100–140 kN preloads in order to achieve
a water-tight sealing, thus allowing water accumulation, which finally led to SCC [156].
Similarly, Neidel et al. [163] reported the failure of an M10 bolt in a gas turbine due to
SCC. In this case, SCC was initiated by pitting corrosion as observed on the bolt. Moreover,
tensile stresses might have been due to three possible reasons. Over-torquing could have
been easily done for these bolts, as the designated torque was 30 Nm, a low value. Secondly,
cold forming was evident from the microstructure, and thirdly, tack welding was performed
to keep the bolt in place. Both of these phenomena induce residual tensile stresses which
contributed to the phenomenon of SCC in the bolt [163]. Hagarova et al. [164] studied the
cause of the failure of a brass bolt used to conduct direct current. The failure was on the
last thread of the bolt, which was characterized by a lower zinc content. SCC was the most
probable cause of failure, as indicated by the dezincification which is associated with this
failure mode.

9.1. Theories of Stress Corrosion Cracking

Efforts had been made to study the underlying mechanisms of SCC. The most common
mechanism is hydrogen embrittlement (HE) [165]. The main factor for hydrogen embrittle-
ment is the presence of hydrogen atoms and a metal. This process reduces the ductility of
the steel by introducing microcracks. The process is initiated by the diffusion of hydrogen
atoms into the lattice of iron, which causes the lattice to inflate. When hydrogen becomes
concentrated enough, cracks start developing. Various sources of hydrogen exist in the
environment which can promote the process, such as water decomposition, the process of
corrosion and by-products of H2 S reaction with iron [166]. Recombination poisons, which
are the adsorbents of hydrogen, control the hydrogen absorption rate. These recombination
poisons act to prevent hydrogen from combining and forming molecules, hence promot-
ing the diffusion of the hydrogen atoms into the iron lattice [167]. The vacancies created
inside the steel by cold working are an easy target for hydrogen accumulation. The atomic
hydrogen diffuses into these sites and accumulates, leading to the HE process [168]. Jha
et al. [169] reported the failure of cadmium-plated 35NCD16 steel (heat-treatable steel with
0.32–0.39% C, min UTS = 1000 MPa, min Sp02 = 800 MPa) threaded fastener because of
hydrogen embrittlement. The most probable reasons behind this occurrence of hydrogen
embrittlement were: (i) spheroidization in an uncontrolled atmosphere, (ii) insufficient
baking, and (iii) a cadmium-plating process which made hydrogen accumulation possible,
plus (iv) residual stresses due to oil quenching [169]. Villalba and Atrens [170] carried
out an experimental study to understand the underlying mechanism of SCC in rock bolts.
They applied accelerated corrosion studies to distinct types of steel used for rock bolt
manufacturing as well as commercial steel alloys. The locations where SCC was observed
were at low pH (2.1) and had negative electrochemical potential. Under this condition
cathodic reaction occurs, forming hydrogen, i.e., 2H + + 2e− → H2 . Hence, they inferred
that hydrogen embrittlement is the most probable cause of failure [170].
Other environmentally assisted degradation mechanisms studied in the literature
comprise the so-called microbiologically influenced corrosion (MIC). In this type of SCC,
anaerobic sulphate-reducing bacteria (SRB) perform sulphur oxidation or sulphate reduc-
tion to generate free sulphuric acids, which can act as a corrosive environment. They
can also form highly corrosive biofilms which can affect the bolts. Moreover, the nitrate-
reducing bacteria (NRB) were found to be a source of SCC in copper. NRB reduce nitrate
into ammonia, which acts as a corrosive environment for copper alloys [171]. They develop
biofilms, acting as corrosive environments at highly localized levels, which contain different
pH values, oxygen and chemical species. The underground regions where rock bolts are
used were found to contain these bacteria [172]. These bacteria, most importantly, produce
hydrogen sulphide, which converts into ferric sulphide reacting with rust. Hydrogen
sulphide also assists in hydrogen adsorption by acting as recombinant poison; hence, SRB
also promotes hydrogen embrittlement [173].
Machines 2023, 11, 212 29 of 41

9.2. Experimental Analyses

Several experimental strategies have been developed to examine stress corrosion crack-
ing based on hydrogen embrittlement as well as microbiologically influenced corrosion.
In 1993, Atrens et al. [174] introduced the use of an experimental apparatus for a linearly
increasing stress test (LIST) method to test SCC. In this test, a load-controlled strategy is
applied to assess the specimen against SCC. In the apparatus, the specimen is loaded at one
end of a lever beam. The other end of the lever beam has a moveable load. This load can be
moved away to increase the moment arm and increase the torque acting on the specimen.
This test has been used by many researchers to assess the SCC on bolts [174]. Gamboa and
Atrens [175] used the LIST apparatus to test various rock bolts, in which rock bolts failed in
a similar manner to the service failure using LIST at a stress rate of 0.019 MPa·s−1 . Villalba
and Atrens [176] also used the LIST testing for testing rock bolts against SCC. They tested
various bolts which failed at a stress rate of 0.019 MPa·s−1 , whereas only five did not fail.
Aziz et al. [177] developed an experimental apparatus to assess the rock bolts subjected
to corrosion conditions with axial loadings of 100 and 200 kN. Water was dripping on the
520–540 mm length of bolts taken from Shoalhaven River inlets (NSW, Australia) with a
pH of 3.4 to 4.3. The experiment was conducted for three and a half years and showed
no initiation of SCC [177]. Vandermaat et al. [178] built a bending and torsion loading
apparatus (BaTLa) to test SCC in rock bolts. They collected mine water from the locations
where SSC failure had occurred. At the time of reporting, the three bolts did not show SSC
after immersion in that environment from 6 months to 1 year [178].
Therefore, removing hydrogen and microorganisms could be regarded as an effective
strategy to mitigate the onset of SCC.

10. Impact Loading

Bolted joints are often subjected to impact loading in different situations in several
applications, such as construction [179], aerospace [180], etc. Therefore, the behaviour of
bolted joints subjected to impact loading must be studied systematically. The influence of
impact loading on threaded fasteners has been studied since 1964. At that time, Hongo [181]
studied the effect of tensile impact on the threaded portion of the fasteners. In 1968,
Koga [182] presented a theory for the mechanism of loosening due to impact loading and
provided solutions according to his theory. He postulated that loosening would occur when
compressive impact occurs on the thread surfaces. The dynamic loading on the nut thread
must be kept greater than or equal to the dynamic loading on the bolt thread to prevent
loosening. The theory was verified with the help of impact experimentation in which the
fastener assembly was lifted to a desired height with the help of four sashcords. The impact
states were measured using strains attached to the apparatus. The experimentation and
hypothesis were in good agreement with each other [182]. Koga and Isono [183] presented
a theoretical model based on self-loosening limit diagrams. Their theory considers the
propagation of compressive and tensile waves when axial impact loading is applied. There
is the repetition of stress waves reaching the free surface of the bolt, which causes the
preload to drop, resulting in the loosening of the bolt [183]. Daadbin and Chow [24]
presented a mass-spring model for studying different parameters affecting the loosening
of bolted joints. Among these parameters, they studied the effect of impact loading on
the joints, which showed that loosening would occur if the spring rebound caused by
the impact is greater than the compression induced by the preload [24]. An analytical
model was presented by Saleem and Nasir [184] for anchor steel bolts loaded by impact.
They studied the effect of impact loading (for pull-out) on the bonding of the anchor
steel bolts using the analytical model and verified it using impact loading imparted by
a Schmidt hammer. The analytical approaches become overly complex for bolted joints
subjected to impact loading, especially when more than one bolt is involved. For this
purpose, researchers have worked to produce models which are less computationally
intensive without much compromise on accuracy [184]. Moore et al. [185,186] proposed
a reduced-order model (ROM) for this purpose and verified its accuracy by means of a
Machines 2023, 11, 212 30 of 41

split Hopkinson bar. The mathematical relationship between joint stiffness and fastener
torque was developed which agreed with the experimental results. The model was able
to predict the preload loss due to the impact loading of the bolted joints. For the impact
velocity of 10.8 ms−1 and 54 Nm initial torque, the model predicted a preload loss of 61.1%,
which was close to the experimentally found preload loss of 63% [185,186]. In another
work, Saleem [187] presented a shear-lag material model for steel anchor bolts subjected
to impact loading. The model considered the cracking at the interface caused by impact
loading along with cyclic pull-out push-in response of bolts. Previous study gave only
impact response of pull-out response of steel anchor bolts. However, both pull-out and
push-in response (cyclic impact loading) is taken into this model, which could be used to
simulate the response of the bolts. The stresses generated with respect to the displacement
of impact loading are found and can be used to check the integrity of bolts as well as the
bolt/concrete interface. The model was in agreement with the numerical simulation as
well as experimental testing [187]. Cao et al. [188] presented a failure criterion for impact
loading on bolted flange joints for aeroengines. They categorized the failure mechanisms
of the bolted flange joints based on the yield or failure of the bolt or flange or both. The
yield and failure modes were then estimated based on the relationship between the plastic
resistances of the flange and the bolt. The experimental validation was performed with
a smoothbore powder gun. It was used to impart impact to the joints at velocities of 90
to 220 ms−1 , which was a representation of debris in a real situation in aero-engines. The
failure modes predicted by the theoretical model were in agreement with the failure modes
obtained through experimentation. Based on the criterion, different design improvements
were suggested by using finite element modelling [188].
Over the years, both experimental and numerical approaches have been used to study
the behaviour of bolted joints subjected to impact loading. Thin panels are joined using
bolts in automobiles which might fail under impact loading due to high plastic strains
induced. Simplified models were introduced which reduced the computational cost without
compromising the accuracy and were verified with one full model. The models were giving
accurate results with significant time savings in model processing [189]. Guo et al. [190]
used the split Hopkinson pressure bar to study the force transmission in bolted flanged
joints under impact loading. The results indicated that higher damping was achieved with
greater values of excitation frequencies. Zheng et al. [191] performed an experimental study
of loosening in threaded connections, still generating impact loadings by means of the
Hopkinson bar. The schematic of the modified Hopkinson bar used in their study is shown
in Figure 21. A high-speed camera was used to capture the loosening angle of the threaded
connection subjected to impact loading. 3D finite element modelling was also performed
considering the single shock wave. For single shock wave assumption, it is necessary to
employ a boundary condition which prevents reflection of the shock wave. For this purpose,
infinite element type (CIND8R in ABAQUS 6.14) was chosen, which enables the setting up
of non-reflective boundaries. The Johnson–Cook model was utilized to simulate accurately
the elevated strain-rate phenomenon. The results showed that loosening depends on initial
preload as well as the amplitude of the stress wave, with every single wave contributing to
the loosening, which accumulates to provide the final loosening.
Furthermore, Ngo et al. [192] studied the impact-loading behaviour of beam-column
joints. They compared the behaviour of steel and GFRP bolts under impact loading using a
pendulum system imparting the impact loading. A load cell captured the applied force,
and the failure of the joint was detected by a high-speed camera, as shown in Figure 22.
The GFRP bolts give the advantage of corrosion resistance compared to the steel joints as
well as preventing the opening of material due to residual stresses. The impact loading
showed shear failure for which more fibres can be applied in order to prevent failure in
that direction.
achines 2023, 11, x FOR PEER REVIEW 31 of

Machines 2023, 11, 212 31 of 41

stress wave, with every single wave contributing to the loosening, which accumulates
provide the final loosening.

Machines 2023, 11, x FOR PEER REVIEW 32 of 42

Figure 21. Apparatus
Figure used
21. Apparatus usedin
in the studyofof
the study impact
impact loading
loading on threaded
on threaded connections
connections [191]. [192].

Furthermore, Ngo et al. [193] studied the impact-loading behaviour of beam-colum

joints. They compared the behaviour of steel and GFRP bolts under impact loading usi
a pendulum system imparting the impact loading. A load cell captured the applied for
and the failure of the joint was detected by a high-speed camera, as shown in Figure 2
The GFRP bolts give the advantage of corrosion resistance compared to the steel joints
well as preventing the opening of material due to residual stresses. The impact loadi
showed shear failure for which more fibres can be applied in order to prevent failure
that direction.

Figure 22. Experimental apparatus for impact testing of beam-column joints [193].
Figure 22. Experimental apparatus for impact testing of beam-column joints [192].
11.Conclusions
11. Conclusions
Severaltypes
Several typesof of failure
failure modes
modes characteristic
characteristic of threaded
of threaded fasteners
fasteners havereviewed.
have been been re-
viewed. From the existing research, it can be deduced that self-loosening
From the existing research, it can be deduced that self-loosening plays a crucial roleplays a crucial
in
role in the failure of threaded connections. This phenomenon is attributable
the failure of threaded connections. This phenomenon is attributable to different causes to different
causes
and and adequate
requires requires adequate
predictivepredictive
modelling.modelling. Specifically,
Specifically, the literaturethereview
literature review
highlights
highlights the need for analytical modelling for rotational loosening as well as for local
slippage accumulation in case of transverse excitation. Moreover, extensive research
should be dedicated to investigating the role of material combinations in loosening.
The body of the literature shows that the uncertainty on the initial preload is the main
cause of failure arising from the assembling phase. Particularly, the correct choice of the
Machines 2023, 11, 212 32 of 41

the need for analytical modelling for rotational loosening as well as for local slippage
accumulation in case of transverse excitation. Moreover, extensive research should be
dedicated to investigating the role of material combinations in loosening.
The body of the literature shows that the uncertainty on the initial preload is the main
cause of failure arising from the assembling phase. Particularly, the correct choice of the
bolting sequence is essential for the correct application of the initial preload in case of
multiple bolted joints. This fosters the development of new efficient algorithms to reduce
the dispersion of the clamping force of multiple bolted joints.
Another key point emerging from this review is the need for a proper material choice
depending on the specific application. This choice must consider the effects of the manu-
facturing process on the texture and microstructure of the material. Close attention must
also be paid to the interaction between the material of the bolt and clamped parts. For
example, differences in the coefficients of thermal expansion could result in the failure of
bolted joints for small temperature variations. Future research direction might consider
the development of new stiffer, lighter and corrosion-resistant bolt materials and their
compatibility with innovative materials used for clamped parts.
The research on new bolt materials must include the development of TSR-CSR dia-
grams, which have been demonstrated to be a fundamental tool to assess fretting fatigue
failures. The development of new coatings to reduce the risk of this kind of failure is
another promising direction for future studies.
The reviewed studies highlighted several factors affecting the fatigue failure of bolted
joints, such as the preload, geometry of the bolts, load spectrum, use of threadlockers and
material processing techniques. Since threadlockers and lubricants have been demonstrated
to significantly improve the fatigue life of bolted joints, new formulations should be inves-
tigated in the future. Another promising research direction is the design of new thread ge-
ometries to reduce the stress concentration and increase the lifespan of threaded fasteners.
It has also been discussed how the incorrect design of the bolted joints with respect
to the applied loadings may lead to failures due to overloading. This phenomenon is also
closely related to the frictional characteristics of the joints. Specifically, lubrication plays a
key role in determining relations between applied torque and preload. In the future, the ef-
fects of anti-seize materials must be further investigated to clarify whether their application
should be limited to the threaded portion or extended to the whole body of the bolts.
As far as environmental factors are concerned, the literature highlighted a predom-
inant role of Stress Corrosion Cracking (SCC) in the failure of threaded joints. The main
mechanisms of SCC include hydrogen embrittlement (HE) and microbiologically influ-
enced corrosion (MIC). The presence of hydrogen must be monitored closely during the
material manufacturing and the service of the bolted joints. To date, there is still very
little knowledge about the MIC of threaded fasteners. The experimental investigation and
analytical modelling of this phenomenon and its inhibitors are thus important issues for
future research.
Many models have been proposed in the literature to explain the process of bolt failure
due to impact loading. The main gap in this area is the absence of an analytical model for
the hexagonal head bolts subjected to impact loading.
From the present research, it can be concluded that, even if the root cause of failure
can often be ascribed to a single mechanism, different mechanisms usually combine to
bring about the final failure of the joint, leaving the actual reason for failure initiation
hard to determine. For instance, if the loosening mechanism of a threaded fastener has
not been correctly accounted for during the design phase, the joint may eventually fail
due to unexpected fatigue loadings after an initial loss of preload has developed. If the
quantity of hydrogen in the environment is beyond the threshold for activation of the SCC,
the fastener can fail abruptly. Similarly, if the heat treatment is not conducted properly,
undesirable metallurgical structures obtained by the material can lead to the initiation
of a crack which can propagate through stress corrosion cracking, overloading or, more
frequently, fatigue. In light of what is discussed above, the designer must account for all
Machines 2023, 11, 212 33 of 41

failure modes applicable, based on the boundary conditions under which the fastener is
intended to operate, and put in place adequate countermeasures. In order to assist the
designer in these considerations, Appendix A provides a short summary of the main types
of failure, along with the relevant causes and some possible mitigation strategies.

Author Contributions: Conceptualization, D.C., M.D.A., S.F., M.M., G.O., C.S. and M.H.B.T.; method-
ology, D.C., M.D.A., S.F., M.M., G.O., C.S. and M.H.B.T.; validation, D.C., M.D.A., S.F., M.M., G.O.,
C.S. and M.H.B.T.; formal analysis, D.C., M.D.A., S.F., M.M., G.O., C.S. and M.H.B.T.; investigation,
D.C., M.D.A., S.F., M.M., G.O., C.S. and M.H.B.T.; resources, D.C., M.D.A., S.F., M.M., G.O., C.S. and
M.H.B.T.; data curation, D.C., M.D.A., S.F., M.M., G.O., C.S. and M.H.B.T.; writing—original draft
preparation, D.C., M.D.A., S.F., M.M., G.O., C.S. and M.H.B.T.; writing—review, editing, D.C., M.D.A.,
S.F., M.M., G.O., C.S. and M.H.B.T.; visualization, D.C., M.D.A., S.F., M.M., G.O., C.S. and M.H.B.T.;
supervision, D.C., M.D.A., S.F., M.M., G.O., C.S. and M.H.B.T.; project administration, D.C., M.D.A.,
S.F., M.M., G.O., C.S. and M.H.B.T.; funding acquisition, D.C., M.D.A., S.F., M.M., G.O., C.S. and
M.H.B.T. All authors have read and agreed to the published version of the manuscript.
Funding: The PhD scholarship of Mr. Tariq has been financed by the National Research and Innova-
tion Operational Program 2014-2020 (CCI 2014IT16M2OP005), ESF REACT-EU resources, Action IV.4
“Doctorates and research contracts on innovation issues” and Action IV.5 Doctorates on green topics.
Scholarship ID DOT1832288-1, CUP J35F21003020006.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: Not applicable.
Conflicts of Interest: The authors declare no conflict of interest.

Appendix A

Table A1. Types of failures of threaded fasteners and their causes.

Failure Type Common Causes Preventions

Use of threadlockers
Vibrational loading exceeding design limits.
Loosening Application of adequate torque during assembly
Preload loss due to any other factor
Use of mechanical retention devices
Clearly state the proper bolting sequences for assembly
of joints having more than one bolt in the
installation drawings.
Incorrect Measuring the initial torque application to avoid the
Application of improper initial preload.
assembly overtightening of bolts.
Carry out experimentation to characterize the friction
coefficient of the joint.
Provide technicians with suitable instruments.
Using rolled threads to have an additional compressive
Allowing high stress concentration regions such as stress layer, which prevents crack formation.
sharp radius of head to shank, the sharp thread Using bolts with smoother head-to-shank transition.
Improper root radius. Rounded thread roots and standardized thread shape
design Manufacturing non-standardized shapes of bolt parts to provide proper fitting of the threads.
parameters such as bolt thread shape. Periodic replacement of bolts.
Unknown changes in friction coefficients upon first Use of appropriate lubrication.
tightening and retightening operations. Carry out experimentation to characterize the friction
coefficient of the joint.
Machines 2023, 11, 212 34 of 41

Table A1. Cont.

Failure Type Common Causes Preventions

Use high corrosion-resistant material such as stainless
steel in humid environments.
Carry out heat treatment in insulated environments
Uncontrolled match between material specifications
especially for sensitive materials such as titanium.
Material and actual components
Refer to relevant standards to define adequate safety
properties Improper heat treatment
factors for specific material combinations, such as
Use of unsuitable lubricants
stainless steel and aluminium.
For critical applications, set up a control strategy for
inspection of the supplied parts.
Removing high stress concentration regions by design.
Fretting Sliding frictional contacts characterized by high
Using appropriate adhesives between
fatigue stress concentration.
matching surfaces.
Apply appropriate initial preload.
Decrease bolt stiffness (material of
Inappropriate geometry of bolted connections. cross-section modification).
Fatigue Improper design. Apply appropriate surface treatments
failure Material defects. (e.g., shot peening).
Inadequate initial preload. Provide accurate design of the most critical parts of the
fastener (1st engaged thread, runout thread,
head-to-shank transition).
Uncontrolled preload (unsuitable torque wrench, or Controlled application of lubricant, only after suitable
uncalibrated torque wrench). experimentation of COF on the same combination
Unsuitable type and application of lubricant. of materials.
Overload
Loading on the system greater than the design loads or Thorough estimation of the loading conditions during
error in estimating the external loads during the design phase (use standards, load cycles from
the design phase. previous products, experimentation on prototypes).
Using materials which are resistant to
hydrogen embrittlement.
Removing hydrogen by employing proper
Presence of hydrogen in the environment and a suitable material processing
Corrosion
material in which it could be adsorbed (e.g., steel) Using dry raw material for steel production with
and other en-
Presence of sulphate-reducing bacteria (SRB) high-quality processes such as vacuum melting
vironmental
Presence of corners during cold working, inducing Allowing enough time in solidification to free up the
effects
stress concentration entrapped hydrogen
Using appropriate coatings on the surfaces of the bolts
Controlling the production of H2 S.
Use of microorganism inhibitors in bolted connections

References
1. Chen, J.-H.; Hsieh, S.-C.; Lee, A.-C. The failure of threaded fasteners due to vibration. Proc. Inst. Mech. Eng. Part C J. Mech. Eng.
Sci. 2005, 219, 299–314. [CrossRef]
2. Haynes, G. Fasteners: A beginner’s guide. Cyber Press: Santa Clara, CA, USA.
3. Milan, M.T.; Spinelli, D.; Filho, W.W.B.; Montezuma, M.F.V.; Tita, V. Failure analysis of a SAE 4340 steel locking bolt. Eng. Fail.
Anal. 2004, 11, 915–924. [CrossRef]
4. Avellon, G. Fastener Failures Prove Surprisingly Common in the Automotive Industry. 2014. Available online: https://boltfailure.
com/fastener-failures-prove-surprisingly-common-in-the-automotive-industry/ (accessed on 5 December 2022).
5. Epower Metals. Failure Analysis of Bolt Corrosion. 2020. Available online: https://www.epowermetals.com/failure-analysis-of-
bolt-corrosion.html (accessed on 5 December 2022).
6. QC-FIT Evaluation of Connector and Bolt Failures—Summary of Findings. 2014. Available online: https://www.bsee.gov/sites/
bsee.gov/files/ (accessed on 5 December 2022).
7. Bickford, J.H. An Introduction to the Design and Behavior of Bolted Joint Marcel Dekker; Raven Press: New York, NY, USA, 1995.
8. Camillo, J. Why Threaded Fasteners Fail. Assembly. 2018. Available online: https://www.assemblymag.com/articles/94240-
why-threaded-fasteners-fail (accessed on 6 December 2022).
9. Pai, N.G. Vibration Induced Loosening of Threaded Fasteners: Mechanisms, Modeling and Design Guidelines. University of
South Florida: Tampa, FL, USA, 2002.
Machines 2023, 11, 212 35 of 41

10. Olmi, G. An experimental investigation on a crack propagating from a geartrain housing in an asphalt milling machine. Eng. Fail.
Anal. 2014, 38, 38–48. [CrossRef]
11. Croccolo, D.; De Agostinis, M.; Fini, S.; Olmi, G. A numerical and experimental approach to the design and failure analysis of a
pinion shaft for wheel loaders. Proc. Inst. Mech. Eng. Part C J. Mech. Eng. Sci. 2018, 232, 1493–1504. [CrossRef]
12. Otter, C.D.; Maljaars, J. Preload loss of stainless steel bolts in aluminium plated slip resistant connections. Thin-Walled Struct.
2020, 157, 106984. [CrossRef]
13. Kharmanda, G.; Olhoff, N.; El-Hami, A. Optimum values of structural safety factors for a predefined reliability level with
extension to multiple limit states. Struct. Multidiscip. Optim. 2004, 27, 421–434. [CrossRef]
14. Wileman, J.; Choudhury, M.; Green, I. Computation of Member Stiffness in Bolted Connections. J. Mech. Des. Trans. ASME 1991,
113, 432–437. [CrossRef]
15. Bhattacharya, A.; Sen, A.; Das, S. An investigation on the anti-loosening characteristics of threaded fasteners under vibratory
conditions. Mech. Mach. Theory 2010, 45, 1215–1225. [CrossRef]
16. Sakai, T. Investigations of bolt loosening mechanisms. Bull. JSME 1978, 21, 1391–1394. Available online: http://www.mendeley.
com/research/geology-volcanic-history-eruptive-style-yakedake-volcano-group-central-japan/ (accessed on 15 December 2022).
[CrossRef]
17. Junker, G.H. New Criteria for Self-Loosening of Fasteners Under Vibration. SAE Tech. Pap. 1969, 78, 314–335. [CrossRef]
18. Yamamoto, A.; Kasei, S. Investigation on the self loosening of threaded fasteners under transverse vibration-A solution for
self-loosening mechanism. J. Jpn. Soc. Precis. Eng. 1977, 43, 470–475. [CrossRef]
19. Pai, N.; Hess, D. Three-dimensional finite element analysis of threaded fastener loosening due to dynamic shear load. Eng. Fail.
Anal. 2002, 9, 383–402. [CrossRef]
20. Li, Z.; Chen, Y.; Sun, W.; Jiang, P.; Pan, J.; Guan, Z. Study on self-loosening mechanism of bolted joint under rotational vibration.
Tribol. Int. 2021, 161, 107074. [CrossRef]
21. Bretl, J.L.; Cook, R.D. Modelling the load transfer in threaded connections by the finite element method. Int. J. Numer. Methods
Eng. 1979, 14, 1359–1377. [CrossRef]
22. Miller, D.L.; Marshek, K.M.; Naji, M.R. Determination of load distribution in a threaded connection. Mech. Mach. Theory 1983, 18,
421–430. [CrossRef]
23. Daadbin, A. Static and dyamic analysis of knuckle shape threaded joints. Mech. Mach. Theory 1990, 25, 225–232.
24. Daadbin, A.; Chow, Y. A theoretical model to study thread loosening. Mech. Mach. Theory 1992, 27, 69–74. [CrossRef]
25. Casanova, F. Failure analysis of the draft tube connecting bolts of a Francis-type hydroelectric power plant. Eng. Fail. Anal. 2009,
16, 2202–2208. [CrossRef]
26. Nassar, S.A.; Housari, B.A. Effect of Thread Pitch and Initial Tension on the Self-Loosening of Threaded Fasteners. J. Press. Vessel.
Technol. Trans. ASME 2005, 128, 590–598. [CrossRef]
27. Housari, B.A.; Nassar, S.A. Effect of Thread and Bearing Friction Coefficients on the Vibration-Induced Loosening of Threaded
Fasteners. J. Vib. Acoust. Trans. ASME 2007, 129, 484–494. [CrossRef]
28. Yang, X.; Nassar, S. Analytical and Experimental Investigation of Self-Loosening of Preloaded Cap Screw Fasteners. J. Vib. Acoust.
2011, 133, 031007. [CrossRef]
29. Nassar, S.A.; Yang, X. A Mathematical Model for Vibration-Induced Loosening of Preloaded Threaded Fasteners. J. Vib. Acoust.
Trans. ASME 2009, 131, 021009. [CrossRef]
30. Nassar, S.A.; Yang, X. Modeling the effect of nut thread profile angle on the vibration-induced loosening of bolted joint systems.
In Advanced Joining Processes; Öchsner, A., da Silva, L.F.M., Altenbach, H., Eds.; Elsevier Inc.: Amsterdam, The Netherlands, 2020;
pp. 29–54.
31. Liu, Z.; Wang, B.; Li, Y.; Zhang, C.; Wang, Y.; Chu, H. Analysis of self-loosening behavior of high strength bolts based on accurate
thread modeling. Eng. Fail. Anal. 2021, 127, 105541. [CrossRef]
32. Goodier, J.N. Loosening by Vibration of Threaded Fastenings. Mech. Eng. 1945, 67, 798.
33. Sauer, J.A.; Lemmon, D.C.; Lynn, E.K. Bolts: How to prevent their loosening. Mach. Des. 1950, 22, 133–139.
34. Paland, E.G. Investigation of the Locking Features of Dynamically Loaded Bolted Connections. Doctoral Dissertation, Hannover
University, Hannover, Germany, 1966.
35. Zaki, A.M.; Nassar, S.A.; Yang, X. Effect of Conical Angle and Thread Pitch on the Self-Loosening Performance of Preloaded
Countersunk-Head Bolts. J. Press. Vessel. Technol. Trans. ASME 2012, 134, 021210. [CrossRef]
36. DIN 65151:2002-08; Aerospace Series—Dynamic Testing of the Locking Characteristics of Fasteners under Transverse Loading
Conditions (Vibration Test). Deutsche Institut für Normung E.V. (DIN): Berlin, Germany, 2002.
37. Dinger, G. Design of multi-bolted joints to prevent self-loosening failure. Proc. Inst. Mech. Eng. Part C J. Mech. Eng. Sci. 2016, 230,
2564–2578. [CrossRef]
38. Dinger, G. and Friedrich, G. Avoiding self-loosening failure of bolted joints with numerical assessment of local contact state. Eng.
Fail. Anal. 2011, 18, 2188–2200. [CrossRef]
39. Chen, Y.; Gao, Q.; Guan, Z. Self-Loosening Failure Analysis of Bolt Joints under Vibration considering the Tightening Process.
Shock. Vib. 2017, 2017, 2038421. [CrossRef]
40. Du, J.; Qiu, Y.; Wang, Z.; Li, J.; Wang, H.; Wang, Z.; Zhang, J. A three-stage criterion to reveal the bolt self-loosening mechanism
under random vibration by strain detection. Eng. Fail. Anal. 2022, 133, 105954. [CrossRef]
Machines 2023, 11, 212 36 of 41

41. Baubles, R.C.; Mccormick, G.J.; Faroni, C.C. Loosening of Fasteners by Vibration; Report No. ER272-2177; Elastic Stop Nut Corp. Am.
(ESNA): Union, NJ, USA, 1966.
42. Zhang, M.; Jiang, Y.; Lee, C.-H. An Experimental Investigation of the Effects of Clamped Length and Loading Direction on
Self-Loosening of Bolted Joints. J. Press. Vessel. Technol. Trans. ASME 2005, 128, 388–393. [CrossRef]
43. Miller, M.; Johnson, C.; Sonne, N.; Mersch, J.; Kuether, R.J.; Smith, J.; Ortiz, J.; Castelluccio, G.; Moore, K. Bolt Preload Loss Due to
Modal Excitation of a C-Beam Structure. In Nonlinear Structures & Systems; Springer: New York, NY, USA, 2020; pp. 203–214.
[CrossRef]
44. Eraliev, O.M.U.; Zhang, Y.-H.; Lee, K.-H.; Lee, C.-H. Experimental investigation on self-loosening of a bolted joint under cyclical
temperature changes. Adv. Mech. Eng. 2021, 13, 1–8. [CrossRef]
45. Casanova, F.; Mantilla, C. Fatigue failure of the bolts connecting a Francis turbine with the shaft. Eng. Fail. Anal. 2018, 90, 1–13.
[CrossRef]
46. Fonte, M.; Reis, L.; Infante, V.; Freitas, M. Failure analysis of cylinder head studs of a four stroke marine diesel engine. Eng. Fail.
Anal. 2019, 101, 298–308. [CrossRef]
47. Jawwad, A.K.A.; Alshabatat, N.; Mahdi, M. The effects of joint design, bolting procedure and load eccentricity on fatigue failure
characteristics of high-strength steel bolts. Eng. Fail. Anal. 2021, 122, 105279. [CrossRef]
48. O’Brien, M.J.; Metcalfe, R.G. High Strength Engineering Fasteners: Design for Fatigue Resistance. J. Fail. Anal. Prev. 2009, 9,
171–181. [CrossRef]
49. Van Campen, D.H. A Systematic Bolt-Tightening Procedure for Reactor Vessel Flanges; University of Technology: Delft, The Nether-
lands, 1969.
50. Bibel, G.D. Method of Tightening Threaded Fasteners. U.S. Patent No. 5,278,775„ 11 January 1994.
51. Nassar, S.A.; Wu, Z.; Yang, X. A New Tightening Methodology for Gasketed Joints Based on Nonlinear Finite Element Analysis.
Am. Soc. Mech. Eng. Press. Vessel. Pip. Div. PVP 2008, 2, 127–137. [CrossRef]
52. Abasolo, M.; Aguirrebeitia, J.; Avilés, R.; De Bustos, I.F. Methodology for the Optimization of Bolting Sequences for Wind
Generator Flanges. J. Press. Vessel. Technol. Trans. ASME 2014, 136, 061202. [CrossRef]
53. Zheng, M.; Li, Y.; Liu, Z.; Zhao, Y.; Yang, C. A multi-bolt tightening method based on elastic interaction coefficient and gasket
creep relaxation. Proc. Inst. Mech. Eng. Part C J. Mech. Eng. Sci. 2022, 236, 9639–9647. [CrossRef]
54. Shiba, R.; Chao, E.Y.S.; Kasman, R. Fatigue Properties of the Hoffmann-Vidal External Fixation Apparatus. Orthopedics 1984, 7,
443–456. [CrossRef] [PubMed]
55. Curtis, L.R. Metallurgical failure analysis of fasteners in an impeller assembly. J. Fail. Anal. Prev. 2006, 6, 17–22. [CrossRef]
56. Eraslan, O.; Inan, O. The effect of thread design on stress distribution in a solid screw implant: A 3D finite element analysis. Clin.
Oral Investig. 2009, 14, 411–416. [CrossRef] [PubMed]
57. Alemayehu, D.-B.; Jeng, Y.-R. Three-Dimensional Finite Element Investigation into Effects of Implant Thread Design and Loading
Rate on Stress Distribution in Dental Implants and Anisotropic Bone. Materials 2021, 14, 6974. [CrossRef] [PubMed]
58. Stevenson, M.; McDougall, J.; Cline, K. Metallurgical Failure Analysis of Titanium Wing Attachment Bolts. Pract. Fail. Anal. 2003,
3, 75–80. [CrossRef]
59. Verein Deutscher Ingenieure (VDI). Systematic Calculation of High Duty Bolted Joints Joints with One Cylindrical Bolt; VDI 2230;
Verein Deutscher Ingenieure (VDI): Düsseldorf, Germany, 2003.
60. Croccolo, D.; De Agostinis, M.; Vincenzi, N. Failure analysis of bolted joints: Effect of friction coefficients in torque–preloading
relationship. Eng. Fail. Anal. 2011, 18, 364–373. [CrossRef]
61. Croccolo, D.; De Agostinis, M.; Vincenzi, N. Influence of tightening procedures and lubrication conditions on titanium screw
joints for lightweight applications. Tribol. Int. 2012, 55, 68–76. [CrossRef]
62. De Agostinis, M.; Fini, S.; Olmi, G. The influence of lubrication on the frictional characteristics of threaded joints for planetary
gearboxes. Proc. Inst. Mech. Eng. Part C J. Mech. Eng. Sci. 2016, 230, 2553–2563. [CrossRef]
63. Han, X.; Thrush, S.J.; Zhang, Z.; Barber, G.C.; Qu, H. Tribological characterization of ZnO nanofluids as fastener lubricants. Wear
2020, 468–469, 203592. [CrossRef]
64. Kumar, M.; Persson, E.; Sherrington, I.; Glavatskih, S. Changes in friction of zinc flake coated threaded fasteners due to humidity,
temperature and storage duration. Tribol. Int. 2022, 170, 107498. [CrossRef]
65. Rezvani, S.; Zhou, L.; Ni, C. Experimental evaluation of angle bracket connections in CLT structures under in- and out-of-plane
lateral loading. Eng. Struct. 2021, 244, 112787. [CrossRef]
66. Croccolo, D.; De Agostinis, M.; Vincenzi, N. Recent improvements and design formulae applied to front motorbike suspensions.
Eng. Fail. Anal. 2010, 17, 1173–1187. [CrossRef]
67. Chen, W.-H.; Lee, S.-S.; Yeh, J.-T. Three-dimensional contact stress analysis of a composite laminate with bolted joint. Compos.
Struct. 1995, 30, 287–297. [CrossRef]
68. Riccio, A. Effects of Geometrical and Material Features on Damage Onset and Propagation in Single-lap Bolted Composite Joints
under Tensile Load: Part II—Numerical Studies. J. Compos. Mater. 2005, 39, 2091–2112. [CrossRef]
69. Valenza, A.; Fiore, V.; Borsellino, C.; Calabrese, L.; Di Bella, G. Failure Map of Composite Laminate Mechanical Joint. J. Compos.
Mater. 2007, 41, 951–964. [CrossRef]
70. Xiao, Y.; Ishikawa, T. Bearing strength and failure behavior of bolted composite joints (part I: Experimental investigation). Compos.
Sci. Technol. 2005, 65, 1022–1031. [CrossRef]
Machines 2023, 11, 212 37 of 41

71. Chang, F.-K.; Lessard, L.B. Damage Tolerance of Laminated Composites Containing an Open Hole and Subjected to Compressive
Loadings: Part I—Analysis. J. Compos. Mater. 1991, 25, 2–43. [CrossRef]
72. Chang, F.-K.; Chang, K.-Y. A Progressive Damage Model for Laminated Composites Containing Stress Concentrations. J. Compos.
Mater. 1987, 21, 834–855. [CrossRef]
73. Dano, M.-L.; Gendron, G.; Picard, A. Stress and failure analysis of mechanically fastened joints in composite laminates. Compos.
Struct. 2000, 50, 287–296. [CrossRef]
74. Camanho, P.P.; Matthews, F.L. Stress analysis and strength prediciton of mechanically fastened joints in FRP: A review. Compos.
Part A Appl. Sci. Manuf. 1997, 28, 529–547. [CrossRef]
75. Hashin, Z. Fatigue Failure Criteria for Unidirectional Fiber Composites. J. Appl. Mech. 1981, 48, 846–852. [CrossRef]
76. Olmedo, Á.; Santiuste, C. On the prediction of bolted single-lap composite joints. Compos. Struct. 2012, 94, 2110–2117. [CrossRef]
77. Camanho, P.; Lambert, M. A design methodology for mechanically fastened joints in laminated composite materials. Compos. Sci.
Technol. 2006, 66, 3004–3020. [CrossRef]
78. Camanho, P.P.; Matthews, F.L. A Progressive Damage Model for Mechanically Fastened Joints in Composite Laminates. J. Compos.
Mater. 1999, 33, 2248–2280. [CrossRef]
79. Hart-Smith, L.J. Bolted Joints in Graphite-Epoxy Composites; Mcdonnell Douglas Corp Long Beach CA Douglas Aircraft DIV: Long
Beach, CA, USA, 1976.
80. ASTM D 7248/D 7248M-08; Standard Test Method for Bearing/Bypass Interaction Response of Polymer Matrix Composite
Laminates Using 2-Fastener Specimens. American Society for Testing and Materials (ASTM) International: West Conshohocken,
PA, USA, 2008.
81. Liu, F.; Zhao, L.; Mehmood, S.; Zhang, J.; Fei, B. A modified failure envelope method for failure prediction of multi-bolt composite
joints. Compos. Sci. Technol. 2013, 83, 54–63. [CrossRef]
82. Cheng, X.; Wang, S.; Zhang, J.; Huang, W.; Cheng, Y.; Zhang, J. Effect of damage on failure mode of multi-bolt composite joints
using failure envelope method. Compos. Struct. 2017, 160, 8–15. [CrossRef]
83. Zhao, L.; Shan, M.; Liu, F.; Zhang, J. A probabilistic model for strength analysis of composite double-lap single-bolt joints. Compos.
Struct. 2017, 161, 419–427. [CrossRef]
84. Davidson, P.; Waas, A.M. Probabilistic defect analysis of fiber reinforced composites using kriging and support vector machine
based surrogates. Compos. Struct. 2018, 195, 186–198. [CrossRef]
85. Liu, F.; Fang, Z.; Zhao, L.; Zhang, J.; Hu, N. A failure-envelope-based method for the probabilistic failure prediction of composite
multi-bolt double-lap joints. Compos. Part B Eng. 2019, 172, 593–602. [CrossRef]
86. ISO 898-1; Mechanical Properties of Fasteners Made of Carbon Steel and Alloy Steel—Part 1: Bolts, Screws and Studs with
Specified Property Classes—Coarse Thread and Fine Pitch Thread. ISO (International Organization of Standardization): London,
UK, 2018.
87. Lee, H.-C.; Jin, Y.-G.; Lee, Y.-H.; Son, I.-H.; Lee, D.-L.; Im, Y.-T. Process design of high-strength bolt of fully pearlitic high-carbon
steel. J. Mater. Process. Technol. 2010, 210, 1870–1875. [CrossRef]
88. Kodur, V.; Yahyai, M.; Rezaeian, A.; Eslami, M.; Poormohamadi, A. Residual mechanical properties of high strength steel bolts
subjected to heating-cooling cycle. J. Constr. Steel Res. 2017, 131, 122–131. [CrossRef]
89. Hu, Y.; Tang, S.-L.; George, A.K.; Tao, Z.; Wang, X.-Q.; Thai, H.-T. Behaviour of stainless steel bolts after exposure to elevated
temperatures. J. Constr. Steel Res. 2019, 157, 371–385. [CrossRef]
90. Ban, H.; Yang, Q.; Shi, Y.; Luo, Z. Constitutive model of high-performance bolts at elevated temperatures. Eng. Struct. 2021, 233,
111889. [CrossRef]
91. Cai, W.-Y.; Jiang, J.; Li, G.-Q.; Wang, Y.-B. Fracture behavior of high-strength bolted steel connections at elevated temperatures.
Eng. Struct. 2021, 245, 112817. [CrossRef]
92. Li, D.; Uy, B.; Wang, J.; Song, Y. Assessment of titanium alloy bolts for structural applications. Steel Compos. Struct. 2022, 42,
553–568. [CrossRef]
93. Adeoti, G.O.; Fan, F.; Huihuan, M.; Shen, S. Investigation of aluminium bolted joint (HBJ) system behavior. Thin-Walled Struct.
2019, 144, 106100. [CrossRef]
94. Cui, Y.; Meng, Q.; Zhang, B.; Yan, F. Fracture analysis of Ti–6Al–4V bolts. Eng. Fail. Anal. 2006, 13, 669–672. [CrossRef]
95. Jha, A.K.; Singh, S.K.; Kiranmayee, M.S.; Sreekumar, K.; Sinha, P. Failure analysis of titanium alloy (Ti6Al4V) fastener used in
aerospace application. Eng. Fail. Anal. 2010, 17, 1457–1465. [CrossRef]
96. Gong, Y.; Ding, Q.; Yang, Z.-G. Failure analysis on premature fracture of anchor bolts in seawater booster pump of nuclear power
plant. Eng. Fail. Anal. 2018, 97, 10–19. [CrossRef]
97. Grimmer, P.W.; Mersch, J.P.; Smith, J.; Veytskin, Y.; Susan, D. Modeling Empirical Size Relationships on Load-Displacement
Behavior and Failure in Threaded Fasteners. In Proceedings of the AIAA Scitech 2019 Forum, San Diego, CA, USA, 7–11 January
2019; p. 2268. [CrossRef]
98. Qi, Y.; Zhang, X.; Shan, L.; Jiao, G.; Gao, Q.; Lin, Z.; Zheng, W.; Zhao, S. Design and fabrication of isotropic high-strength Ti-6Al-4V
fastener. Mater. Lett. 2022, 323, 132554. [CrossRef]
99. Radouani, R.; Echcharqy, Y.; Essahli, M. Numerical Simulation of Galvanic Corrosion between Carbon Steel and Low Alloy Steel
in a Bolted Joint. Int. J. Corros. 2017, 2017, 6174904. [CrossRef]
Machines 2023, 11, 212 38 of 41

100. Zhang, J.; Ebrahimi, N.; Lai, D. Galvanic Corrosion Risk of Using Galvanized A325 Bolts in Corrosion-Resistant Steel Bridges. J.
Bridg. Eng. 2019, 24, 06019001. [CrossRef]
101. Abid, M.; Al-Grafi, M.W.; Wajid, H.A. Effect of bolt behaviour on the performance of a bolted joint. World Appl. Sci. J. 2013, 26,
89–96. [CrossRef]
102. Lacalle, R.; Cicero, S.; Ferreño, D.; Álvarez, J. Failure analysis of a bolt in a scaffolding system. Eng. Fail. Anal. 2008, 15, 237–246.
[CrossRef]
103. Korsunsky, A.M. (Ed.) Chapter 11—Eigenstrain Methods in Structural Integrity Analysis. In A Teaching Essay on Residual Stresses
and Eigenstrains; Elsevier Inc.: Cambridge, MA, USA, 2017; pp. 167–172.
104. Walker, P.B. Fretting in the Light of Aircraft Experience. J. R. Aeronaut. Soc. 1959, 63, 293–298. [CrossRef]
105. Sandifer, J.P. Evaluation of methods for reducing fretting fatigue damage in 2024-T3 aliminium lap joints. Wear 1973, 26, 405–412.
[CrossRef]
106. Wagle, S.; Kato, H. Ultrasonic detection of fretting fatigue damage at bolt joints of aluminum alloy plates. Int. J. Fatigue 2009, 31,
1378–1385. [CrossRef]
107. Chakherlou, T.; Razavi, M.; Aghdam, A.; Abazadeh, B. An experimental investigation of the bolt clamping force and friction
effect on the fatigue behavior of aluminum alloy 2024-T3 double shear lap joint. Mater. Des. 2011, 32, 4641–4649. [CrossRef]
108. Croccolo, D.; De Agostinis, M.; Fini, S.; Olmi, G.; Robusto, F.; Scapecchi, C. Fretting fatigue in mechanical joints: A literature
review. Lubricants 2022, 10, 53. [CrossRef]
109. Ungar, E.E. Energy Dissipation at Structural Joints; Mechanisms and Magnitudes; Bolt Beranek and Newman Inc.: Cambridge, MA,
USA, 1964. [CrossRef]
110. Rogers, P.F.; Boothroyd, G. Damping at Metallic Interfaces Subjected to Oscillating Tangential Loads. J. Eng. Ind. 1975, 97,
1087–1093. [CrossRef]
111. Eriten, M.; Polycarpou, A.A.; Bergman, L.A. Development of a Lap Joint Fretting Apparatus. Exp. Mech. 2011, 51, 1405–1419.
[CrossRef]
112. Gaul, L.; Lenz, J. Nonlinear dynamics of structures assembled by bolted joints. Acta Mech. 1997, 125, 169–181. [CrossRef]
113. Ovcharenko, A.; Halperin, G.; Etsion, I.; Varenberg, M. A novel test rig for in situ and real time optical measurement of the contact
area evolution during pre-sliding of a spherical contact. Tribol. Lett. 2006, 23, 55–63. [CrossRef]
114. Kartal, M.E.; Mulvihill, D.M.; Nowell, D.; Hills, D.A. Determination of the Frictional Properties of Titanium and Nickel Alloys
Using the Digital Image Correlation Method. Exp. Mech. 2010, 51, 359–371. [CrossRef]
115. Hartwigsen, C.; Song, Y.; McFarland, D.; Bergman, L.; Vakakis, A. Experimental study of non-linear effects in a typical shear lap
joint configuration. J. Sound Vib. 2004, 277, 327–351. [CrossRef]
116. Smallwood, D.O.; Gregory, D.L.; Coleman, R.G. A three parameter constitutive model for a joint which exhibits a power law
relationship between energy loss and relative displacement. In Proceedings of the 72nd Shock and Vibration Symposium, Destin,
FL, USA, 12–16 November 2001; pp. 12–16.
117. Mutoh, Y.; Jayaprakash, M. Tangential stress range–compressive stress range diagram for fretting fatigue design curve. Tribol. Int.
2011, 44, 1394–1399. [CrossRef]
118. Murugesan, J.; Mutoh, Y. Fretting fatigue strength prediction of dovetail joint and bolted joint by using the generalized tangential
stress range–compressive stress range diagram. Tribol. Int. 2014, 76, 116–121. [CrossRef]
119. Juoksukangas, J.; Lehtovaara, A.; Mäntylä, A. Experimental and numerical investigation of fretting fatigue behavior in bolted
joints. Tribol. Int. 2016, 103, 440–448. [CrossRef]
120. Jiménez-Peña, C.; Talemi, R.H.; Rossi, B.; Debruyne, D. Investigations on the fretting fatigue failure mechanism of bolted joints in
high strength steel subjected to different levels of pre-tension. Tribol. Int. 2017, 108, 128–140. [CrossRef]
121. Li, D.; Xu, C.; Botto, D.; Zhang, Z.; Gola, M. A fretting test apparatus for measuring friction hysteresis of bolted joints. Tribol. Int.
2020, 151, 106431. [CrossRef]
122. Szlosarek, R.; Kröger, M. Fatigue behavior of bolted boreholes under various preloads. Mater. Test. 2022, 64, 195–201. [CrossRef]
123. Ferjaoui, A.; Yue, T.; Wahab, M.A.; Hojjati-Talemi, R. Prediction of fretting fatigue crack initiation in double lap bolted joint using
Continuum Damage Mechanics. Int. J. Fatigue 2015, 73, 66–76. [CrossRef]
124. Aghababaei, R.; Warner, D.H.; Molinari, R.A.J.-F. Critical length scale controls adhesive wear mechanisms. Nat. Commun. 2016, 7,
11816. [CrossRef] [PubMed]
125. Leonard, B.D.; Ghosh, A.; Sadeghi, F.; Shinde, S.; Mittelbach, M. Third body modeling in fretting using the combined finite-discrete
element method. Int. J. Solids Struct. 2014, 51, 1375–1389. [CrossRef]
126. Rodríguez-Tembleque, M.H.A.L.; Abascal, R. Anisotropic Fretting Wear Simulation Using the Boundary Element Method. Comput.
Model. Eng./Sci. 2012, 87, 127–156. [CrossRef]
127. McColl, I.R.; Ding, J.; Leen, S.B. Finite element simulation and experimental validation of fretting wear. Wear 2004, 256, 1114–1127.
[CrossRef]
128. Armand, J.; Pesaresi, L.; Salles, L.; Wong, C.; Schwingshackl, C.W. A modelling approach for the nonlinear dynamics of assembled
structures undergoing fretting wear. Proc. R. Soc. A Math. Phys. Eng. Sci. 2019, 475, 20180731. [CrossRef] [PubMed]
129. Li, X.; Yang, J. Fatigue Behavior of Bolted Joints: Comparative Analysis for the Effect of Fretting and Stress Concentration. Mater.
Perform. Charact. 2020, 9, 309–323. [CrossRef]
Machines 2023, 11, 212 39 of 41

130. Antaki, G.; Gilada, R. (Eds.) Chapter 2—Design Basis Loads and Qualification. In Nuclear Power Plant Safety and Mechanical
Integrity; Butterworth-Heinemann: Boston, MA, USA, 2015; pp. 27–102.
131. Garcia, T.; Cicero, S.; Ibáñez, F.; Álvarez, J.; Martín-Meizoso, A.; Bannister, A.; Klimpel, A.; Aldazabal, J. Fatigue Performance of
Thermally Cut Bolt Holes in Structural Steel S460M. Procedia Eng. 2015, 133, 590–602. [CrossRef]
132. Croccolo, D.; Bogojević, N.; De Agostinis, M.; Fini, S.; Olmi, G.; Robusto, F.; Šoškić, Z.; Ćirić-Kostić, S. Fatigue response of
additively manufactured as-built 15-5 PH stainless steel and effects of machining and thermal and surface treatments. Fatigue
Fract. Eng. Mater. Struct. 2022, 46, 433–451. [CrossRef]
133. Fini, S.; Croccolo, D.; De Agostinis, M.; Olmi, G.; Robusto, F.; Scapecchi, C. Experimental Investigation on the Fatigue Strength for
Different Tightening Procedures and Materials in Metric Screws. Press. Vessel. Pip. Conf. 2022, 86151, V002T02A031.
134. Oskouei, R.; Chakherlou, T. Reduction in clamping force due to applied longitudinal load to aerospace structural bolted plates.
Aerosp. Sci. Technol. 2009, 13, 325–330. [CrossRef]
135. Albrecht, P.; Sahli, A.H.; Wattar, F. Fatigue Strength of Bolted Joints. Eng. Struct. 1987, 113, 1834–1849. [CrossRef]
136. Huhn, H.; Valtinat, G. Bolted connections with hot dip galvanized steel members with punched holes. In Proceedings of the
ECCS/AISC Workshop, Connections in Steel Structures V: Innovative Steel Connections, Amsterdam, The Netherlands, 3–4 June
2004; pp. 3–5.
137. Chakherlou, T.N.; Oskouei, R.; Vogwell, J. Experimental and numerical investigation of the efect of clamping force on the fatigue
behaviour of bolted plates. Eng. Fail. Anal. 2008, 15, 563–574. [CrossRef]
138. Esmaeili, F.; Chakherlou, T.; Zehsaz, M. Investigation of bolt clamping force on the fatigue life of double lap simple bolted and
hybrid (bolted/bonded) joints via experimental and numerical analysis. Eng. Fail. Anal. 2014, 45, 406–420. [CrossRef]
139. Esmaeili, F.; Barzegar, S.; Jafarzadeh, H. Evaluation of fatigue life reduction factors at bolt hole in double lap bolted joints using
volumetric method. J. Solid Mech. 2017, 9, 1–11.
140. Abazadeh, B. Failure mode prediction of single lap bolted Joints. Int. Res. J. Adv. Eng. Sci. 2016, 1, 95–99.
141. Munse, W.H.; Wright, D.T.; Newmark, N.M. Laboratory Tests of Bolted Joints. Trans. Am. Soc. Civ. Eng. 1955, 120, 1299–1318.
[CrossRef]
142. Griza, S.; da Silva, M.E.G.; dos Santos, S.V.; Pizzio, E.; Strohaecker, T.R. The effect of bolt length in the fatigue strength of M24×3
bolt studs. Eng. Fail. Anal. 2013, 34, 397–406. [CrossRef]
143. Birkemoe, P.C.; Meinheit, D.; Munse, W.H. Fatigue of a 514 steel in bolted connections. J. Struct. Div. 1969, 95, 2011–2030.
[CrossRef]
144. Atzori, B.; Lazzarin, P.; Quaresimin, M. A re-analysis on fatigue data of aluminium alloy bolted joints. Int. J. Fatigue 1997, 19,
579–588. [CrossRef]
145. Zhou, S.; Sun, Y.; Ge, J.; Chen, X. Multiaxial fatigue life prediction of composite bolted joint under constant amplitude cycle
loading. Compos. Part B Eng. 2015, 74, 131–137. [CrossRef]
146. Vogt, J.-B.; Boussac, O.; Foct, J. Prediction of fatigue resistance of a hot-dip galvanized steel. Fatigue Fract. Eng. Mater. Struct. 2001,
24, 33–39. [CrossRef]
147. Cicero, S.; García, T.; Álvarez, J.; Martín-Meizoso, A.; Aldazabal, J.; Bannister, A.; Klimpel, A. Definition and validation of
Eurocode 3 FAT classes for structural steels containing oxy-fuel, plasma and laser cut holes. Int. J. Fatigue 2016, 87, 50–58.
[CrossRef]
148. Dragoni, E. Effect of anaerobic threadlockers on the fatigue strength of threaded connections. Int. J. Mater. Prod. Technol. 1999, 14,
445. [CrossRef]
149. Taylor, D.S.; Pan, J. Fatigue Crack Initiation in Threaded Fasteners Using the Critical Plane Approach. SAE Trans. 2003, 112,
318–325. [CrossRef]
150. Hareyama, S.; Manabe, K.; Shimodaira, T.; Hoshi, A. Working load analysis and strength estimation for bolted joints. In
Proceedings of the ASME 2014 International Mechaical Engineering Congress and Exposition (IMECE2014), Montreal, QC,
Canada, 14–20 November 2014; pp. 1–10.
151. Sorg, A.; Utzinger, J.; Seufert, B.; Oechsner, M. Fatigue life estimation of screws under multiaxial loading using a local approach.
Int. J. Fatigue 2017, 104, 43–51. [CrossRef]
152. Nassar, S.A.; Li, T. Novel Modeling of Mean Stress Effect on the High-Cycle Fatigue Performance of Preloaded Threaded Fasteners.
Am. Soc. Mech. Eng. Press. Vessel. Pip. Div. (Publ.) PVP 2019, 2, 1–7. [CrossRef]
153. Baggerly, R.G. Evaluating the Effects of Overtorque in Bolts. J. Fail. Anal. Prev. 2001, 1, 41–46. [CrossRef]
154. Eliaz, N.; Gheorghiu, G.; Sheinkopf, H.; Levi, O.; Shemesh, G.; Ben-Mordechai, A.; Artzi, H. Failures of bolts in helicopter main
rotor drive plate assembly due to improper application of lubricant. Eng. Fail. Anal. 2003, 10, 443–451. [CrossRef]
155. Xue, X.; Tang, X.; Lu, H.; Geng, X. Failure Analysis of a Reactor after Explosion Accident. Procedia Eng. 2015, 130, 1397–1403.
[CrossRef]
156. Chen, H.; Ramandi, H.L.; Walker, J.; Crosky, A.; Saydam, S. Failure of the threaded region of rockbolts in underground coal mines.
Min. Technol. 2018, 127, 146–154. [CrossRef]
157. Neidel, A.; Gädicke, T.; Ullrich, T. Torsional Overload Fracture of Twist-off Bolts During Assembly. Pract. Metallogr. 2022, 59,
39–49. [CrossRef]
Machines 2023, 11, 212 40 of 41

158. Vandermaat, D.; Elias, E.; Craig, P.; Saydam, S.; Crosky, A. Experimental protocol for stress corrosion cracking of rockbolts. In
Proceedings of the 2012 Coal Operators’ Conference, Mining Engineering, University of Wollongong, Wollongong, Austrilia,
16–17 February 2012; pp. 129–136.
159. Craig, P.; Saydam, S.; Hagan, P.; Crosky, A.; Hebblewhite, B. Australian rock bolt steel chemistry and stress corrosion cracking. In
Proceedings of the Second Australasian Ground Control in Minig Conference, Sydney, NSW, Australia, 23–24 November 2010; pp.
40–44.
160. Crosky, A.; Smith, B.; Hebblewhite, B. Failure of rockbolts in underground mines in Australia. J. Fail. Anal. Prev. 2003, 3, 70–78.
[CrossRef]
161. Escobar, J.; Romero, A.; Lobo-Guerrero, J. Failure analysis of submersible pump system collapse caused by assembly bolt crack
propagation by stress corrosion cracking. Eng. Fail. Anal. 2015, 60, 1–8. [CrossRef]
162. Craig, P.H.; Hagan, P.; Vandermaat, D. Premature failure of australian coal mine rock bolts. Int. J. Min. Sci. Technol. 2016, 26,
59–64. [CrossRef]
163. Neidel, A.; Gädicke, T.; Hartanto, V.; Wallich, S.; Wöhl, E. Austenitic Stainless Steel Bolt Failure by Stress Corrosion Cracking.
Pract. Metallogr. 2018, 55, 97–109. [CrossRef]
164. Hagarová, M.; Peterka, P.; Mantič, M.; Vojtko, M.; Baranová, G.; Matvija, M. Failure analysis of leaded brass bolt. Eng. Fail. Anal.
2023, 143, 106899. [CrossRef]
165. Perrin, M.; Gaillet, L.; Tessier, C.; Idrissi, H. Hydrogen embrittlement of prestressing cables. Corros. Sci. 2010, 52, 1915–1926.
[CrossRef]
166. Ćwiek, J. Prevention methods against hydrogen degradation of steel. J. Achiev. Mater. Manuf. Eng. 2010, 43, 214–221.
167. Biezma, M. The role of hydrogen in microbiologically influenced corrosion and stress corrosion cracking. Int. J. Hydrogen Energy
2001, 26, 515–520. [CrossRef]
168. Toribio, J.; Valiente, A. Failure analysis of cold drawn eutectoid steel wires for prestressed concrete. Eng. Fail. Anal. 2006, 13,
301–311. [CrossRef]
169. Jha, A.K.; Narayanan, P.R.; Sreekumar, K.; Mittal, M.; Ninan, K. Hydrogen embrittlement of 3.5Ni–1.5Cr–0.5Mo steel fastener.
Eng. Fail. Anal. 2008, 15, 431–439. [CrossRef]
170. Villalba, E.; Atrens, A. Hydrogen embrittlement and rock bolt stress corrosion cracking. Eng. Fail. Anal. 2009, 16, 164–175.
[CrossRef]
171. Al-Nabulsi, K.M.; Al-Abbas, F.M.; Rizk, T.Y.; Salameh, A.E.M. Microbiologically assisted stress corrosion cracking in the presence
of nitrate reducing bacteria. Eng. Fail. Anal. 2015, 58, 165–172. [CrossRef]
172. AlAbbas, F.M.; Williamson, C.; Bhola, S.M.; Spear, J.R.; Olson, D.L.; Mishra, B.; Kakpovbia, A.E. Influence of sulfate reducing
bacterial biofilm on corrosion behavior of low-alloy, high-strength steel (API-5L X80). Int. Biodeterior. Biodegrad. 2013, 78, 34–42.
[CrossRef]
173. Chen, H.; Kimyon, Ö.; Ramandi, H.L.; Craig, P.; Gunawan, C.; Wu, S.; Manefield, M.; Crosky, A.; Saydam, S. Microbiologically
influenced stress corrosion cracking responsible for catastrophic failure of cable bolts. Eng. Fail. Anal. 2022, 131, 105884. [CrossRef]
174. Atrens, A.; Brosnan, C.C.; Ramamurthy, S.; Oehlert, A.; O Smith, I. Linearly increasing stress test (LIST) for SCC research. Meas.
Sci. Technol. 1993, 4, 1281–1292. [CrossRef]
175. Gamboa, E.; Atrens, A. Environmental influence on the stress corrosion cracking of rock bolts. Eng. Fail. Anal. 2003, 10, 521–558.
[CrossRef]
176. Villalba, E.; Atrens, A. An evaluation of steels subjected to rock bolt SCC conditions. Eng. Fail. Anal. 2007, 14, 1351–1393.
[CrossRef]
177. Aziz, N.; Craig, P.; Nemcik, J.; Hai, F. Rock bolt corrosion—An experimental study. In 2013 Coal Operators’ Conference; Mining
Engineering, University of Wollongong: Wollongong, Australia, 2013; pp. 144–151.
178. Vandermaat, D.; Saydam, S.; Hagan, P.; Crosky, A. Examination of rockbolt stress corrosion cracking utilising full size rockbolts in
a controlled mine environment. Int. J. Rock Mech. Min. Sci. 2016, 81, 86–95. [CrossRef]
179. Grimsmo, E.L.; Clausen, A.; Aalberg, A.; Langseth, M. Beam-to-column joints subjected to impact loading. In Proceedings of the
Nordic Steel Construction Conference, Tampere, Finland, 23–25 September 2015.
180. Mohotti, D.; Ali, M.; Ngo, T.; Lu, J.; Mendis, P.; Ruan, D. Out-of-plane impact resistance of aluminium plates subjected to low
velocity impacts. Mater. Des. 2013, 50, 413–426. [CrossRef]
181. Hongo, K. Loosening of Bolt and Nut Fastenings. Trans. Jpn. Soc. Mech. Eng. 1964, 30, 934–939. [CrossRef]
182. Koga, K. Loosening by Repeated Impact of Threaded Fastenings. Trans. Jpn. Soc. Mech. Eng. 1969, 35, 1104–1111. [CrossRef]
183. Koga, K.; Isono, H. Study on Self-loosening of Bolted Joints Taking Account of Characteristics of Impulsive Friction. Bull. JSME
1986, 29, 1004–1012. [CrossRef]
184. Saleem, M.; Nasir, M. Bond evaluation of steel bolts for concrete subjected to impact loading. Mater. Struct. 2015, 49, 3635–3646.
[CrossRef]
185. Moore, K.J. A Reduced-Order Model for Loosening of Bolted Joints Subjected to Axial Shock Excitation. J. Appl. Mech. 2019, 86,
1–37. [CrossRef]
186. Moore, K.J.; Kurt, M.; Eriten, M.; Dodson, J.C.; Foley, J.R.; Wolfson, J.C.; McFarland, D.M.; Bergman, L.A.; Vakakis, A.F. Nonlinear
Parameter Identification of a Mechanical Interface Based on Primary Wave Scattering. Exp. Mech. 2017, 57, 1495–1508. [CrossRef]
Machines 2023, 11, 212 41 of 41

187. Saleem, M. Cyclic Shear-Lag Model of Steel Bolt for Concrete Subjected to Impact Loading. J. Mater. Civ. Eng. 2018, 30, 04018023.
[CrossRef]
188. Cao, Z.; Zhang, F.; Zhang, D.; Yu, Y.; Li, L.; Guo, X. Failure mechanisms of bolted flanges in aero-engine casings subjected to
impact loading. Chin. J. Aeronaut. 2021, 34, 125–144. [CrossRef]
189. Tanlak, N.; Sonmez, F.; Talay, E. Detailed and simplified models of bolted joints under impact loading. J. Strain Anal. Eng. Des.
2011, 46, 213–225. [CrossRef]
190. Guo, Y.; Wei, Y.; Yang, Z.; Huang, C.; Wu, X.; Yin, Q. Nonlinearity of interfaces and force transmission of bolted flange joints
under impact loading. Int. J. Impact Eng. 2017, 109, 214–223. [CrossRef]
191. Zheng, Q.; Guo, Y.; Wei, Y.; Wang, Y.; Wang, X. Loosening of steel threaded connection subjected to axial compressive impact
loading. Int. J. Impact Eng. 2020, 144, 103662. [CrossRef]
192. Ngo, T.T.; Pham, T.M.; Hao, H.; Chen, W.; Elchalakani, M. Performance of monolithic and dry joints with GFRP bolts reinforced
with different fibres and GFRP bars under impact loading. Eng. Struct. 2021, 240, 112341. [CrossRef]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.