Fig. 1. Schematic shows of (a): load-time curves of CRL test with various strain rate, (b) an example of load/ displacement-time curves of CLH creep test.

Fig. 2. XRD patterns of Cu-Mg alloyed thin films.

Fig. 3. (a1-c1) Microstructure evolution with different Mg composition. The top right corner is corresponding SADPs. Growth direction is marked with blue arrows. (a2-c2) Corresponding HRTEM images of twin morphologies in samples with different Mg composition. (a3-c3) Grain size d histograms of Cu-Mg alloyed thin films with different composition.

Fig. 4. Statistic data of grain size d and average twin spacing λ as well as twin fraction PT are obtained by fitting the data via the Lognormal function of corresponding histograms. As is indicated by the red lines in (a) average grain size; black and blue lines in (b) twin spacing and fraction as a function of Mg composition, respectively.

Fig. 5. An overview of one of 3DAP samples of Cu-6.5 at.% Mg containing two GBs. (a) Front view and top view of a needle-like sample. (b) Mg atoms distribution along blue tubularis direction in Fig. 5(a). Red solid line is the Gaussian compositional field fitting of Mg concentration curve. Segregational width δ of the two GBs is 5.03 nm and 4.07 nm, respectively.

Fig. 6. Post mortem TEM features of the indented Cu-6.5 at.% Mg films. (a) The cross-sectional TEM images after nanoindentation test. Here we defined three positions with different deformation degree. Position 1: undeformed area; 2: medium deformed area; 3: severely deformed area. (b) A magnified, bright-field TEM image of the area near indentation. (c) The corresponding dark-field TEM image of the deformed area in (b), shows grain apex barreling under indenter during creep test. (d) Grain sizes of these three positions. Inserted images: 3D-SADPs taken from: left-up corner: position 1 and right-down corner: position 3.

Fig. 7. Post mortem TEM features of the indented Cu-16.8 at.% Mg films. (a) The cross-sectional TEM images after nanoindentation test. Here we defined three positions with different deformation degree. Position 1: undeformed area; 2: medium deformed area; 3: severely deformed area. (b) A magnified, bright-field TEM image of the area near indentation. (c) The corresponding dark-field TEM image of the severely deformed area in (b). Inserted HRTEM image shows deformation twins adjoining indented surface. (d) Grain sizes of these three positions. Inserted images: 3D-SADPs taken from: left-up corner: position 1 and right-up corner: position 3.

Fig. 8. Constant rate of load (CRL) tests. (a) An example of load-depth curves of Cu-2.8 at.% Mg sample in the CRL test with different strain rate. (b) Hardness of Cu-Mg alloyed thin films as well as pure Cu is plotted as a function of strain rates. The slope of linear fit represents corresponding strain rate sensitivity (SRS m).

Fig. 9. Constant load and hold (CLH) creep tests. (a) Representative load-depth curves of various Cu-Mg alloyed thin films. (b) Representative depth-time experimental data and fitting curve of Cu-6.5 at.% Mg sample. (c) Creep response curves of Cu films alloyed with different Mg composition. Inserted image shows creep depth of various samples with different Mg composition. (d) An example of the experimental method to determine the strain rate sensitivity (SRS mc) in CLH test.

Fig. 10. Strain rate sensitivity (SRS m) values of Cu-Mg alloyed thin films derived from both CRL test and CLH creep test. As a comparison, the strain rate sensitivity of coarse grained, bulk Cu-Mg alloy with a Mg composition of ～ 6 at.% is also measured. The SRS m in CRL test and SRS mc in CLH creep test are plotted by half-hollow symbols in red and blue, respectively.

Fig. 11. Nanoindentation results of elastic modulus and hardness and corresponding fitting curves: (a) the measured elastic modulus of Cu-Mg (Mg spanning from 0-16.8 at.%) samples and linear fitting curve and a comparison with the reported values of polycrystalline pure Cu in literature [30,60]; (b) the calculated two types of solution-related strengthening mechanisms, containing Fleischer strengthening and R-S softening; (c) the calculated reduced GB energy as a function of Mg concentration at GB; (d) the measured hardness (stress) and a combined strengthening model to quantify the Mg concentration-dependent hardness of NT Cu-Mg thin films.

Fig. 12. Comparison of SRS m derived from different testing methods of the present Cu-Mg alloyed thin films and the reported NC Cu [30,40,68,69], NT Cu [20,23,70] as well as bulk Cu [20], is plotted as a function of grain size. Pure Cu and Cu alloys with nanotwins (marked as NT) are plotted as solid symbols. Pure Cu with non-twins (marked as NC) are plotted as half-hollow symbols.