1

Peening effects on fatigue Crack Growth in Friction Stir Welds

Omar Hatamleh1,*, Jed Lyons2 , & Royce Forman3

1) Structures & Dynamics Branch, NASA-Johnson Space Center, Houston, Texas 77058 2) Mechanical Engineering Dept., University of South Carolina, Columbia, South Carolina 29208 3) Materials & Processes Branch, NASA- Johnson Space Center, Houston, Texas 77058 Abstract The surface modification from laser and shot peening was used to introduce compressive

residual stresses into friction stir welded (FSW) Aluminum Alloy (AA) 7075-T7351.

Their influence on the fatigue crack growth of FSW was characterized and evaluated for two

different crack configurations. The results indicate a significant decrease in fatigue crack

growth rates resulting from using laser peening compared to shot peening versus their native

welded specimens.

Keywords: FSW, laser peening, shot peening, fatigue crack growth 1. Introduction

Friction stir welding (FSW) (illustrated in Figure 1) is a relatively new welding

technique invented by the Welding Institute in England in 1991 [1]. Since then, this

technique has emerged as a promising solid state process with encouraging results,

especially when used on high strength aerospace aluminum alloys that are usually

difficult to weld [2]. Using frictional heating from a rotating tool, combined with forging

pressure, high strength bonds are produced at temperatures below the melting

temperature of the material [3]. * Corresponding author. Tel: (281) 483-0286; Fax: (281) 244-5918 Email: omar.hatamleh-1@nasa.gov

2

An adequate spindle tilt toward the trailing direction ensures that the tool shoulder

holds the stirred material by the threaded pin, and the material is moved efficiently from

the front of the pin to the back.

Figure 1 Principle of the friction stir welding process.

Residual stresses in FSW may be considerably less than those in fusion welds since

FSW takes place at a lower temperature than fusion welding. However the rigid

clamping arrangement used in FSW, along with the heating cycle the material

experiences during welding, can still significantly affect residual stresses in the FSW

weld [4, 5]. These residual stresses can significantly affect the service performance of

welded materials by facilitating the fatigue crack growth process [6]. Residual tensile

stresses in welds can also lead to faster crack initiation.

The use of FSW is becoming more popular due to the advantages it offers compared

to other conventional fusion welding techniques. The capability of FSW to weld high

strength aluminum alloys like AA 7000 series has resulted in welded joints being used in

Advancing Side

Retreating Side

Rotating Tool

3

critical load bearing structures, and is being used by modern industries for structurally

demanding applications [7]. This has instigated the need for techniques and methods that

can alleviate the tensile residual stresses in welded components.

Several studies [8-19] have investigated the fatigue behavior of FSW aluminum

alloys, but none of these studies has focused on the effects of laser peening on fatigue

crack growth. Laser peening (LP) (shown in Figure 2) is a technique with the capability

to introduce a state of residual compressive stresses that can significantly increases

fatigue properties [20, 21].

Figure 2 Laser peening process

The LP process utilizes high energy laser pulses (several GW/cm2) fired at the

surface of a metal coated with an ablative film, and covered with a transparent layer

(usually water). As the laser beam passes through the transparent layer and hits the

surface of the material, a thin layer of the ablative layer is vaporized. The vapor

continues to absorb the remaining laser energy and is heated and ionized intoa plasma.

The rapidly expanding plasma is trapped between the sample and the transparent layer,

creating a high surface pressure, which propagates into the material as a shock wave [22].

4

When the peak pressure of the shock wave is greater than the dynamic yield

strength of the material, it produces extensive plastic deformation in the metal. The actual

depths of the LP induced stresses will vary depending on the type of material, the laser

peening processing conditions chosen, and the material properties [23]. Compressive

stresses produced by laser peening can also be generated deeper below the surface by

using successive shocks. Previous research [24, 25] has shown that the residual stress

resulting from laser peening can be significantly higher and deeper than for conventional

shot peening.

In this study, the surface modification from laser and shot peening was used to

introduce compressive residual stresses into FSW AA 7075-T7351. Changes in fatigue

crack growth from differences in residual stresses resulting from different peening

techniques was assessed and evaluated. The effects of crack orientation with respect to

the weld have also been explored.

2. Experimental Procedure

The aluminum alloy 7075-T651 was used in this investigation. AA 7075 is a

precipitation-hardened aluminum alloy widely used in aerospace applications due to its

high strength. The 7075-T651 was supplied as a 6.35mm plate with an ultimate and yield

strength of 561 MPa, and 536 MPa, respectively, and an elongation of 11%.

The FSW specimens for this investigation were made at the NASA Johnson Space

Center. The rotational speed used to weld the plates was 350 RPM in the

counterclockwise direction, and the translation speed was 2.54 cm/min. The welding

direction was aligned with the rolling direction. The FSW panels produced by NASA

5

were 122 cm x 40 cm x 0.65 cm. Following the welding process, the welded plates were

aged from the T651 condition to the T7351 condition. The FSW puts the weld nugget

microstructure in a supersaturated solid solution condition; therefore, heat treatment is

necessary to prevent the welded material from continuing to age at room temperature [26,

27].

Following the heat treatment, the welded plates were inspected using radiographic

and penetrant inspections. The inspections results did not reveal any indication of voids

or defects in the weld. After that, bending tests using strip specimens with dimensions of

17.8 cm x 2.54 cm were done. Both the root and the crown sides of the weld were tested

to evaluate the quality of the weld. The samples were inspected visually afterward with

no crack indications revealed.

The mechanical properties of the welded plates were verified by a tension test.

The tensile specimens used in the test consisted of conventional dog bone coupons in

accordance with ASTM 08 using specimens with a width of 12.7 mm. The orientation of

the specimens was with the weld in the center of the specimen and the load was applied

perpendicular to the weld direction. The ultimate and yield strength of the tested coupons

were 339 MPa and 226 MPa respectively with an elongation of 5.5%.

Optical micrographs were taken of the transverse cross section of a weld. The

specimen was cut and sectioned and then subjected to several successive steps of

grinding and polishing until proper surface finish was achieved for a metallographic

analysis. After that, the specimen was etched using a Keller’s reagent.

6

Before the shot and laser peening were applied, the specimens were milled on the

top side of the weld removing about 0.4 mm of material. The dimensions of the 6mm

thick M(T) specimen and the crack locations are shown in Figure 3. The specimens

dimensions were in accordance with ATMS E 647. Some fatigue crack growth

specimens were shot-peened using 0.0234” glass beads with an Almen intensity of 0.008-

0.02A and a 100% coverage rate. Other specimens were peened using single and triple

layers of laser peening.

The laser peening was performed at the Metal Improvement Company in

Livermore California, and was applied using a square laser spot size of 4.72 x 4.72 mm2

with a laser power density of 4 GW/cm2 and 18ns in duration. The spots within a layer

were overlapped 3%. Peening between layers had an off-set of 50% in the two in plane

directions. A peening frequency of 2.7 Hz and a 1 micron wavelength laser was

employed. Both sides of the specimen were shocked using the same conditions.

After the peening process, a 0.25 mm thick through thickness notch was

introduced to the testing specimens using an electric discharge machine (EDM). The

notch was 12.5 mm in length and was introduced at two different locations. One was at

the plate weld centerline, and the other was 7.5 mm from the edge of the weld at a

location corresponding to the Heat Affected Zone (HAZ).

7

(a) Configuration I (b) Configuration II

Figure 3 Fatigue test coupons configurations (Suggestion to label the samples as

Configuration I and Configuration II to make the later reading easier to follow)

3 Results and discussion

3.1. Weld microstructure and hardness

A cross section of the weld is illustrated in Figure 4. The cross section revealed

the classical formation of the elliptical onion rings structure in the center of the weld.

The FSW samples investigated revealed no visible porosity or defects.

8

Figure 4 A Cross section of the welded specimen

A 50x magnification of the cross section of the weld showing the transition from

the nugget- Thermo Mechanical Affected Zone (TMAZ)-HAZ microstructure on the

retreating area of the weld is illustrated in Figure 5. Relative to the nugget, the TMAZ

experiences a lesser degree of plastic deformation and is exposed to lower temperatures;

therefore, recrystallization is not evident in this region. The grain structure in this region

is elongated, with some considerable distortions that may be attributed to mechanical

action from the welding tool. The HAZ is unaffected by mechanical effects and has a

grain structure that resembles the parent material grain structure. Previous work by Jata

et al [28] has indicated that strengthening precipitates in this region have grown in size

and were several times larger than in the parent material. Transitions from TMAZ to the

HAZ and from the HAZ to the base material are gradual and not distinguished by any

abrupt change in microstructure.

Figure 5 A section of the weld nugget-TMAZ-HAZ interface at 50x

9

Figure 6 reveals the fine and equiaxed grains typical of a recrystallised structure

at different regions of the nugget and fairly uniform grains with no apparent defects. The

grain sizes in this region are of the order of 5-10 μm, and are significantly smaller than

the parent material grain due to the higher temperature and extensive plastic deformation.

It is also noted that the average grain size was different for different regions of the

nugget. For example, the grains at the bottom side of the plate were relatively smaller

than those at the top. This may be attributed to the fact that the bottom surface of the

FSW plate is in contact with the backing plate, which may act as a heat sink. Therefore,

lower temperatures and shorter thermal cycle at the bottom of the welded plate

effectively retards the grain growth and results in smaller grains. These findings confirm

the results obtained by Mahoney et al [29] showing similar results for AA 7050.

Figure 6 Nugget structure at 1000x at different regions of weld nugget

10

A micro-hardness test was also performed on a cross section perpendicular to the

welding direction. The test was conducted on a Struers micro-hardness machine, and the

measurements were taken using a 300g for 3 seconds. The results of the test are

illustrated in Figure 7. The figure shows a softened region corresponding to the weld

nugget. The variations in hardness can be correlated to the microstructure developed after

the welding process. The center of the nugget was significantly harder than the TMAZ

immediately outside the weld nugget boundary. The soft region is probably caused by

coarsening and dissolution of strengthening precipitates during the thermal cycle of the

FSW. Softening was produced throughout the weld zone. The lowest hardness did not lie

in the center of the weld, but around 12 mm away from the weld centerline,

corresponding to the weld interface. The hardness levels increased as precipitation

hardening became more effective with increasing distance from the weld.

Microhardness Profile

0

20

40

60

80

100

120

140

160

180

-29 -27 -25 -23 -21 -19 -17 -15 -13 -11 -9 -7 -5 -3 -1 1 3 5 6 8 10 11 13 15 16 18 20 21 23Millimeters Across Weld

Har

dnes

s R

eadi

ng (K

noop

-300

gf)

Figure 7 Micro-hardness test across the weld of FSW 7075-T7351

Nugget TMAZ TMAZ HAZ HAZ Base Base

11

3.2. Fatigue Crack Propagation

The fatigue testing was performed under axial loading at constant

amplitude using a servo-hydraulic machine. The crack length was measured using the

direct current potential difference method. This technique utilizes a constant DC current

passed through a gage attached to the specimen while the voltage difference across the

notch is monitored. All fatigue crack growth (FCG) specimens measurements were

carried out at constant amplitude loads in laboratory air at a frequency of 22 Hz, and at a

stress ratio R=0.1. Before the FCG tests were conducted, the specimens were cycled

until a pre-crack measuring 0.5 mm was obtained from the EDM notch. The fatigue pre-

crack provided a sharpened fatigue crack and symmetry for the M(T) specimen, which

has an important effect on subsequent crack growth data. Upon fatigue cycling, the crack

propagated almost instantaneously from the notch.

The fatigue behavior in the tests was determined by measuring the fatigue crack

length (a) versus the number of cycles (N) for the base non-welded material, the non-

shocked FSW, shot peened, and two laser peened conditions. Figure 8 shows the crack

propagation test data for configuration one (T-L) in AA 7075-T7351 at R=0.1. If we

consider the fatigue life to be the point where the a-N curves become almost vertical, the

specimen processed with three layers of laser peening had substantial fatigue life

improvements with a fatigue life about 200% longer that the non-peened FSW, and

around 40% higher than the base un-welded material. The shot peened specimens tested

did not result in any measurable decrease to the fatigue crack growth rate. The fatigue

life improvement from laser peening was due to the higher and deeper compressive

residual stress resulting from laser compared to conventional shot peening [24, 25].

12

Figure 8 Results for crack length vs. number of cycles for FSW 7075-T7351

Figure 9 illustrates the fatigue crack growth rates using different peening

techniques. The crack growth rates are monotonically increasing, which suggest that the

applied stress intensity factor is increasing as a function of crack length for a given

applied load. The behavior of the baseline specimen is typical of most conventional

metallic samples, where cracking initiates at the machined notch tip and grows to failure

under continual fatigue cycling [30]. The crack growth rates for the three layer laser

peening were around 51-69% less than the crack growth rates for the unpeened and the

shot peened FSW specimens at ΔK=8.5-11.5 mmMPa .

0

5

10

15

20

25

30

35

40

0 20000 40000 60000 80000 100000 120000 140000Number of Cycles

Cra

ck L

engt

h (m

m)

Laser (3 layers)Laser (1 layer)Shot PeeningNo PeeningBase

13

On the other hand, FCG was around 28-42% less for specimens processed with a

single layer of laser peening. The differences in crack growth rates started to decrease

and results started to converge at ΔK >20 mmMPa . These results are similar to the

one reported by Bussu et al [8] for FSW welds using AA 2024-T351. This trend occurs

because as cracks increase in length, the stresses are relaxed and crack growth rates

reduced. This was further confirmed by Galatolo et al [31] where measured residual

stresses acting on the crack tip decreased when the crack size got larger, eventually

disappearing when the crack length was 40mm. Similar trends were also reported by

Donne et al [17], where base and FSW FCG results were converging at higher stress

ratios. This was attributed to the fact that residual stresses were leveled off by the big

plastic zone ahead of the crack.

It was also noted from Figure 9, that the unpeened FSW specimens had a higher

fatigue crack growth rate when compared to the base material. One possible explanation

could be that the microstructure in the HAZ corresponds to one of an overaged structure.

As discussed previously, while grains in the HAZ are similar to the base material; the

strengthening precipitates in this region have grown in size. The resistance to fatigue

crack growth should decrease in microstructures that are overaged [32]. Therefore,

microstructures in which strengthening precipitates are coarsened usually have higher

FCG rates compared to microstructures that contain fine precipitates.

Nevertheless, John et al [26] offered a different explanation for behavior by

investigating the FCG in the HAZ for specimens with different dimensions, and then

compared the results to the base unwelded material. To ensure that the microstructure in

the vicinity of the crack growth region was similar in all specimens, the crack growth

14

occurred in the same region with respect to the weld. The results indicated that the

difference in FCG between the FSW and the base material was due to the contribution

from the residual stresses, and not the microstructure. The results obtained here are

different than those obtained by Ignat’eva et al [33] for fusion welds. In fusion welding,

crack growth rates in the HAZ were reduced compared to the parent material. This could

be due to different residual stress distributions in different welding processes.

1.00E -05

1.00E -04

1.00E -03

1.00E -02

1 10 100

da/d

n (m

m/c

ycle

)

Laser (3 layers)Laser (1 layer)S hot P eeningNo PeeningBase

Figure 9 Crack growth rates for configuration one in FSW 7075-T7351

The number of cycles to grow a 25mm crack from one side of the EDM notch for

configuration one is shown in Figure 10. While improvement from shot peened over the

unpeened specimens was negligible, laser peening resulted in an improvement of 27%,

and 74% in the number of cycles for one and three laser layers respectively. It is also

noted that specimens processed with one layer of laser peening did not surpass the base

15

material, while specimens processed with three layers still had an improvement of 20%

over the base unwelded material.

0

20000

40000

60000

80000

100000

120000

Base No Peening S hot Peening Laser Peening (1 layer)

Laser Peening (3 layers)

Cyc

less

Figure 10 Number of cycles to grow a 25mm crack from on side of the EDM Notch for Configuration I in FSW 7075-T7351

The FCG data for Configuration II (L-T) Al 7075-T7351 under R=0.1 constant

amplitude loading are shown in Figure 11. The data in the graph compares the base un-

welded material to FSW welded specimen both with no peening and peened using

different techniques and parameters. The shot peened specimen in this study did not

improve the fatigue crack growth. Similar results were identified by Honda et al [34]

were it was revealed that overall, shot peening had negligible or only minor effect on

crack growth rates in AA 7075-T7351.

Specimens processed with three layers of laser peening exhibited the highest

increase in fatigue life with an increase of around 280% over the unpeened FSW

specimens, and around 190% over the base un-welded material. Yang [20] attributed the

reduction in FCG rates in laser peened specimens to the compressive residual stress

resulting from the peening process. As a result, the effective stress intensity factor that

controls the FCG in the peened specimen is lower than that of the unpeened ones.

16

0

5

10

15

20

25

30

35

40

0 20000 40000 60000 80000 100000 120000 140000 160000Number of Cycles

Cra

ck L

engt

h (m

m)

Laser (3 layers)Laser (1 layer)S hot PeeningNo PeeningBase Material

Figure 11 Results for crack length vs. number of cycles for FSW 7075-T7351

Figure 12 illustrates the fatigue crack growth rates for the same specimen using

different peening techniques. The crack growth rates for the three layers of laser peening

were around 55-82% less than the crack growth rates for the unpeened specimen for

ΔK>10 mmMPa . The results between the different specimens started to converge at

ΔK>20 mmMPa . Although the crack growth rates for cracks initiating at the center of

the weld were faster than the base unwelded material, they did not reach crack growth

rates as reported by Bussu et al [8] where it was shown that crack growth rates for cracks

starting at the FSW centerline were approximately 10 times faster than parent material.

17

1.00E -05

1.00E -04

1.00E -03

1.00E -02

1 10 100

da/d

n (m

m/c

ycle

)

Laser (3 layers)

No Peening

Base

Figure 12 Crack growth rates for configuration two in FSW 7075-T7351

The number of cycles to grow a 25mm crack from one side of the EDM notch for

Configuration II is also shown in Figure 13. In this configuration, shot peened specimens

did not result in any improvement over the unpeened ones. On the other hand, laser

peening resulted in an improvement of 123% for one layer, and 217% for three layers of

laser peening. Specimen processed with three layers also had a 91% increase in the

number of cycles over the base unwelded material.

18

0

20000

40000

60000

80000

100000

120000

140000

160000

Base No Peening S hot Peening Laser Peening (1 layer)

Laser Peening (3 layers)

Cyc

les

Figure 13 Number of cycles to grow a 25mm crack from on side of the EDM Notch for configuration two in FSW 7075-T7351

Comparisons between the fatigue crack growth for configuration one and two are

illustrated in Figures 14, and 15. The results show fatigue crack growth differences

between one and three layers of laser peening and the unpeened FSW specimen. The

differences between the unpeened specimen for both Configurations I and II were small,

while, in the case of laser peening, the results exhibited more profound differences

between both configurations.

For example, in the case of three layers of laser peening, the differences between

Configuration I and II were almost identical until a crack length of 12.5 mm was reached.

At that point the benefits in fatigue life for Configuration II surpassed the fatigue life in

Configuration I by 34%. These differences are mostly attributed to residual stress

distributions in the different crack configurations.

19

0

5

10

15

20

25

30

35

40

0 20000 40000 60000 80000 100000 120000 140000 160000 180000Number of Cycles

Cra

ck L

engt

h (m

m)

Laser 300% Conf2Laser 300% Conf1Unpeened Conf2Unpeened Conf1

Figure 14 Differences in fatigue life between configuration one and two for specimens peened with a triple layer of laser peening on FSW 7075-T7351

0

5

10

15

20

25

30

35

40

0 20000 40000 60000 80000 100000 120000Number of Cycles

Cra

ck L

engt

h (m

m)

Laser 100% Conf2Laser 100% Conf1Unpeened Conf2Unpeened Conf1

Figure 15 Differences in fatigue life between configuration one and two for specimens peened with a single layer of laser peening on FSW 7075-T7351

20

3.3. Fractography

To evaluate the FCG behavior, the specimens were loaded to fracture under static

tensile loading. Fatigue striation patterns were clearly visible on the fractured surfaces of

the tested FSW 7075-T7351 specimens. Nevertheless, the fatigue striations spacing for

the laser peened specimens were smaller when compared to the un-peened, and shot

peened specimens. This reduction in striation spacing indicates a slower FCG rate and is

partially attributed to the deeper compressive residual stresses induced by the laser

peening. The fatigue striations for an un-peened ConfigurationI specimen are shown in

Figure 16. These pictures were taken at different locations across the crack.

Figure 16 Fractographic images for the fracture surface for Configuration I for a FSW AA 7075-T7351 specimen

21

Comparisons between fatigue striations for Configuration I are shown in figure 17

for different peening techniques. The graph represents striation spacing at different

locations across the crack. Under constant stress intensity conditions, striation spacing in

a local region can vary by a factor of two to four [35]. The main reason for this scatter is

attributed primarily to the fact that striations formation is a highly localized event. The

striations spacing is also dependent on both the stress intensity factor and metallurgical

factors such as variations in the grain orientation, and the distribution of inclusions in that

particular area. Therefore, several measurements of striation spacing were taken to get a

meaningful estimate of fatigue striations at a particular crack length.

The fatigue striations from the triple layer of laser peening exhibited the least

striation spacing compared to the other investigated specimen. The un-peened condition

resulted in the highest striation spacing in the group. The difference in striation spacing

between the peened and un-peened conditions was reduced at larger crack sizes. This is

attributed to the fact that as the crack increases in size, more residual stresses will be

released; eventually the majority of the residual stresses induced by the peening process

will be diminished at a large crack size, and the peening becomes ineffective in reducing

the FCG process. These results seem to be in agreement with the fatigue crack growth

rates in Figures 9 and 12.

22

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

1mm 5mm 12mmDistance from notch

Stria

tion

Spac

ing

(μm

)UnpeenedS hot P eenedLaser P eened (100% )Laser P eened (300% )

Figure 17 Striation spacing for configuration one at different distances from the

notch

The fatigue striations for Configuration II are also shown in Figure 18. These

pictures were taken at different locations across the weld. The fracture surface

appearance was different according to the location across the weld. In the nugget region,

the fracture surface was flat and smooth. At 15mm from the notch, the surface

morphology was rough. As the fatigue crack extends, the stress intensity factor and the

plastic zone size increase in size. When the plastic zone is large compared to the

thickness of the specimen, plane stress conditions and slant fracture start to take effect.

23

Figure 18 Fractographic images for the fracture surface for configuration two for a FSW AA 7075-T7351

Figure 19 illustrates the fatigue striation spacing for a laser peened FSW specimen

at different distances from the edge of the sample. The striation spacing was higher at

3.0mm compared to the spacing at 0.5 mm from the edge of the specimen. This

difference is attributed to the fact that closer to the edge, the plastic zone is generally

higher because of constraint effects which generate higher resistance to crack growth

rates. The difference may also be attributed to the fact that compressive stresses from the

laser peening are expected to be higher in magnitude at 0.5 mm than at the center of the

crack. Toward the center of the specimen, residual stresses are expected to be tensile to

balance the compressive stresses introduced at the near surface of the specimen.

24

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1mm 2mm 5mm 7mm 12mm 1mm 2mm 5mm 7mm 12mm3 mm from surface

Stria

tion

Spac

ing

(μm

)

=

0.5 mm from surface

Figure 19 Striation spacing for a laser peened specimen at 0.5 and 3 mm from the surface

Fatigue striation spacing measurements along the crack length for specimens

with and without laser peening is presented in Figure 20. Fatigue striations spacing

which can be directly related to the crack growth process were again less than the

unpeened condition. These results seem to agree with the results from [36] where the

effect of compressive stresses for a rivet hole before and after cold working was

investigated in relation to fatigue striation in 2024-T3 aluminum alloy. As in

configuration one, the difference in striation spacing between the peened and un-peened

conditions started to diminish as the crack size increased.

00.10.20.30.40.50.60.70.80.9

1mm 2mm 5mm 7mm 12mmNo Peening

Stria

tion

Spac

ing

(μm

)

=

00.10.20.30.40.50.60.70.80.9

1mm 2mm 5mm 7mm 12mmLaser Peening (3 layers)

Stria

tion

Spac

ing

(μm

)

=

Figure 20 Striation spacing for peened and un-peened FSW 7075-T7351

25

4 Summary

The surface modification from laser and shot peening was used to introduce

compressive residual stresses into friction stir weld (FSW) AA 7075-T7351. The fatigue

behavior in the tests was determined by measuring the fatigue crack length (a) versus the

number of cycles (N) for the base non-welded material, the non-shocked FSW, shot

peened, and two laser peened conditions. Their influence on the fatigue crack growth

rate of FSW was characterized and evaluated for two different crack configurations.

The laser peening using three layers resulted in substantial increase in fatigue life

when compared to the unaltered condition. The improvement ranged between 200-280%

depending on the crack configuration. Fatigue striation spacing for the laser peened

specimen was less than other specimen. This reduction in striation spacing indicates a

slower FCG rate, and was attributed in part to the deeper compressive residual stresses

induced by the laser peening.

5 Acknowledgment

The authors are grateful to Mrs. Irene E. Kaye, Mr. Joseph E. Rogers, and Mr.

Gregory F. Galbreath from the NASA Johnson Space Center for their logistical support

for this project. The authors are also grateful to Mr. Raymond Patin, and John Figert

from the NASA Johnson Space Center, and Dr. Lloyd Hackel from the Metal

Improvement Company for their valuable comments.

26

References

1 Thomas W. M et al. Friction stir butt welding. Int Patent App PCT/GB92/02203,

and GB Patent App 9125978.8, December 1991. US patent No. 5, 460,317,

October 1995

2 J. Q. Sue at al. Microstructural investigation of friction stir welded 7050- T651

aluminum. Acta Materialia 51 (2003) 713–729

3 P. Staron et al. Residual stress in friction stir-welded Al sheets. Physica B 350

(2004) e491-e493

4 Mishra R. S. et al. Friction Stir Welding and Processing. Materials Science and

engineering R 50 (2005) 1-78.

5 Donne, C. D. et al. Investigations on Residual Stresses in Friction Stir Welds.

3’rd International Symposium on Friction Stir Welding, Kobe, Japan 2001.

6 Sutton, M. A., Yang, B., Reynolds, A. P., and Taylor, R., 2000, ‘‘Preliminary

Studies of Mixed Mode Fracture in 2024-T3 Friction Stir Welds,’’ Best of

Aeromat Session, ASM Materials Solutions Conference & Exhibition, St. Louis,

MO, October 9–12

7 Cavaliere, P. Effect of friction stir processing on the fatigue properties of a Zr-

modified 2014 aluminum alloy. Materials Characterization. In press 2006.

8 Bussu G. et al. The Role of Residual Stress and Heat affected Zone Properties on

Fatigue crack propagation in friction stir welded 2024-T351 aluminum joints.

International Journal of Fatigue 25 (2003) 77-88

27

9 Biallas G, Dalle Donne C, Juricic C. Monotonic and cyclic strength of

friction stir welded aluminium joints. In: Miannay D, Costa P, Franc¸ois D,

editors. Advances in mechanical behaviour plasticity and damage. Proceedings of

EUROMAT 2000, vol. 1. Amsterdam: Elsevier; 2000. p. 115–20

10 Magnusson L , Kallman L. Mechanical properties of friction stir welds in thin

sheet of aluminium 2024, 6013 and 7475. Second international symposium on

FSW, Gothenburg, Sweden, June 2000

11 Maddox SJ. Review of fatigue assessment procedures for welded aluminum

structures. Int J Fatigue 2003;25:1359–78

12 James MN, Bradley GR. Weld tool travel speed effects on fatigue life of friction

stir welds in 5083 Aluminium. Int J Fatigue 2003;25: 1389–98

13 Ericsson M, Sandstrom. Influence of welding speed on the fatigue of friction stir

welds and comparison with MIG and TIG. Int J Fatigue 2003;25:1379–87

14 Dickerson TL. Fatigue of friction stir welds in Aluminum alloys that contain root

flaws. Int J Fatigue 2003;25:1399–409

15 Okura I, Naruo M, Vigh LG, Hagisawa N, Toda H. Fatigue of aluminum deck

fabricated by friction stir welding. Eighth international conference INALCO 2001

16 Banes, M. et al. Fatigue properties of as-welded AA6005 and AA6082 aluminum alloys

in T1 and T5 temper condition. Trends in Welding Research, Proceedings of the 4th

International Conference, 5-8 June 1995, Gatlinburg, Tennessee.

17 Donne, C, et al. Fatigue and Fracture performance of friction stir welded 2024-T3 joints.

Proceedings European Conference on Spacecraft Structures, Materials and Mechanical

Testing, Brauschweig, Germany, 4-6 November 1998.

28

18 Russel, S. et al. Static, fatigue and crack growth behavior of friction stir welded 7075-T6

and 2024-T3 aluminum alloys. Friction Stir Welding and Processing. The Minerals,

Metals & Materials Society) 2001. pp 93-104.

19 Donne, C, et al. Effect of weld imperfections and residual stresses on the fatigue crack

propagation in friction stir welded joints. Second International Conference on Friction

Stir Welding, 26-28 June 2000. Gothenburg, Sweeden.

20 J. M. Yang at al. Laser shock peening on fatigue behavior of 2024-T3 Al alloy

with fastener holes and stopholes. Materials Science and Engineering A298

(2001) 296–299

21 C. Rubio-Gonzalez et al. Effect of laser shock processing on fatigue crack growth

and fracture toughness of 6061-T6 aluminum alloy. Materials Science and

Engineering A 386 (2004) 291–295

22 Tan Y. et al. Laser shock peening on fatigue crack growth behavior of aluminum

alloy. Fatigue and Fracture of Engineering Materials. 27, 649-656.

23 J. K. Gregory, H. J. Rack, and D. Eylon (eds.) Surface Performance of Titanium,

TMS, Warrendale, PA. (1996) pp. 217-230

24 C. Montross et al. Laser shock processing and its effects on microstructure and

properties of metal alloys: a review. International Journal of Fatigue 24 (2002)

1021-1036

25 P. Peyre at al. Laser shock processing of aluminium alloys. Application to high

cycle fatigue behaviour. Material Science and Engineering A210. (1996) 102-113

29

26 John, R., et al. Residual stress effects on near-threshold fatigue crack growth in

friction stir welds in aerospace alloys. International Journal of Fatigue 25 (2003)

939-948

27 Mahoney, M. W. et al. Properties of friction –stir-welded 7075 T651 aluminum.

Metallurgical and Materials Transactions A. Volume 29A, 1998. pp 1955-1964.

28 Jata K et al. Metall Materials Transactions. 2000, 31A. 2181-2192.

29 M. Mahoney, R.S. Mishra, T. Nelson, J. Flintoff, R. Islamgaliev, Y. Hovansky,

in: K.V. Jata, M.W. Mahoney, R.S. Mishra, S.L. Semiatin, D.P. Filed (Eds.),

Friction Stir Welding and Processing, TMS, Warrendale, PA, USA, 2001, p.183.

30 Rushau J. et al. Fatigue crack nucleation and growth rate behavior of laser shock peened

titanium. International Journal of Fatigue 21 (1999) S199-S209.

31 Galatolo R. et al. Fatigue crack propagation in residual stress fields of welded plates.

International Journal of Fatigue. Vol. 19, No 1, pp 43-49, 1997.

32 Suresh S. et al. Mechanisms of slow fatigue crack growth in high strength aluminum alloys: role of microstructure and environment. Met Transactions. 1984. 15A, 369-379.

33 Ignat’eva V, et al. Effect of residual stress on the development of fatigue cracks in the

region of butt welds. Automatic Welding 1995. 38(1).

34 Honda T. et al. Effect of shot peening on fatigue crack growth in 7075-T7351. Journal of ASTM International, 2005, Vol. 2, No. 6

35 Richard Hertzberg. Deformation and Fracture Mechanics of Engineering Materials. Fourth edition. 1996. John Wiley & sons, Inc.

36 Matos F. et al. Residual stress effect on fatigue striation spacing in a cold-worked rivet hole. Theoretical and Applied Fracture Mechanics 42 (2004) 139–148.