Limits...
Accelerated search for materials with targeted properties by adaptive design.

Xue D, Balachandran PV, Hogden J, Theiler J, Xue D, Lookman T - Nat Commun (2016)

Bottom Line: Here we show how an adaptive design strategy, tightly coupled with experiments, can accelerate the discovery process by sequentially identifying the next experiments or calculations, to effectively navigate the complex search space.We synthesize and characterize 36 predicted compositions (9 feedback loops) from a potential space of ∼800,000 compositions.Of these, 14 had smaller ΔT than any of the 22 in the original data set.

View Article: PubMed Central - PubMed

Affiliation: Theoretical Division, Los Alamos National Laboratory, MS-B262, Los Alamos, New Mexico 87545, USA.

ABSTRACT
Finding new materials with targeted properties has traditionally been guided by intuition, and trial and error. With increasing chemical complexity, the combinatorial possibilities are too large for an Edisonian approach to be practical. Here we show how an adaptive design strategy, tightly coupled with experiments, can accelerate the discovery process by sequentially identifying the next experiments or calculations, to effectively navigate the complex search space. Our strategy uses inference and global optimization to balance the trade-off between exploitation and exploration of the search space. We demonstrate this by finding very low thermal hysteresis (ΔT) NiTi-based shape memory alloys, with Ti50.0Ni46.7Cu0.8Fe2.3Pd0.2 possessing the smallest ΔT (1.84 K). We synthesize and characterize 36 predicted compositions (9 feedback loops) from a potential space of ∼800,000 compositions. Of these, 14 had smaller ΔT than any of the 22 in the original data set.

No MeSH data available.


Related in: MedlinePlus

Results and insights from inference and global optimization.(a) The experimental measurements for thermal hysteresis ΔT as a function of the number of iterations of the loop of Fig. 1 compared with the predictions (inset). Iteration 0 is the original training set of 22 alloys. At each iteration (from 1 onwards), four new predicted alloys are synthesized. The difference between the predicted and measured values of ΔT is large for iterations 1 and 2, drops significantly for iterations 3–6 and then increases beyond iteration 7. We interpret this as illustrating exploration in the early iterations, finding a reasonable minimum in the middle iterations and then exploring new areas in later iterations. (b) The ΔT as a function of the VEN feature shows that the exploration after iteration 3 is confined to an apparent minimum in the narrow interval (6.9:7; inset), favouring the B2→R transformation that is known to have the smallest ΔT (global minimum) compared with B19 and B19' transformations. (c) The average valence electron number of the four synthesized alloys as a function of the number of iterations, showing the exploratory nature (large standard deviation (s.d.)/error bars during iterations 1–2 and from 7 onwards) of the adaptive design in this feature space. The error bars denote standard deviations for VEN over the four samples. The tenth iteration indicates that the design is drifting away from the apparent global minimum (∼6.96 in the y axis).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4835535&req=5

f3: Results and insights from inference and global optimization.(a) The experimental measurements for thermal hysteresis ΔT as a function of the number of iterations of the loop of Fig. 1 compared with the predictions (inset). Iteration 0 is the original training set of 22 alloys. At each iteration (from 1 onwards), four new predicted alloys are synthesized. The difference between the predicted and measured values of ΔT is large for iterations 1 and 2, drops significantly for iterations 3–6 and then increases beyond iteration 7. We interpret this as illustrating exploration in the early iterations, finding a reasonable minimum in the middle iterations and then exploring new areas in later iterations. (b) The ΔT as a function of the VEN feature shows that the exploration after iteration 3 is confined to an apparent minimum in the narrow interval (6.9:7; inset), favouring the B2→R transformation that is known to have the smallest ΔT (global minimum) compared with B19 and B19' transformations. (c) The average valence electron number of the four synthesized alloys as a function of the number of iterations, showing the exploratory nature (large standard deviation (s.d.)/error bars during iterations 1–2 and from 7 onwards) of the adaptive design in this feature space. The error bars denote standard deviations for VEN over the four samples. The tenth iteration indicates that the design is drifting away from the apparent global minimum (∼6.96 in the y axis).

Mentions: In total, we performed nine design iterations and the results are shown in Fig. 3a, which depicts how the experimental (as well as predicted, inset) ΔT behaves with successive iterations. The range (max–min) in ΔT is large in the first two iterations, becomes smaller in iterations 3–6 and increases from the seventh iteration onwards. Figure 3b shows how the measured ΔT varies with the average VEN (one of the features used in the inference step). There are two local minima in the VEN landscape, one at 6.95 and the other at 7.13, and SVRrbf:KG predominantly explores the minima for the VEN at ∼6.95, which eventually led to the discovery of the Ti50.0Ni46.7Cu0.8Fe2.3Pd0.2 alloy. It can be seen in Fig. 3a,b that from the third iteration onwards, our strategy produces results in the desired direction of minimizing ΔT. However, as shown in Fig. 3c, after the sixth iteration our design drifts away from the apparent global minimum. We found 14 new alloys, out of 36 synthesized compositions from 9 feedback loops (see Supplementary Table 2), with ΔT <3.15 K (the best value in our original training set). One of the alloys, Ti50.0Ni46.7Cu0.8Fe2.3Pd0.2 with B2 to R transformation, discovered in iteration 6 had a ΔT of 1.84 K (as measured from DSC curves), surpassing the best value in the training data by 42%. In Table 1, we list the best five low ΔT alloys from this work.


Accelerated search for materials with targeted properties by adaptive design.

Xue D, Balachandran PV, Hogden J, Theiler J, Xue D, Lookman T - Nat Commun (2016)

Results and insights from inference and global optimization.(a) The experimental measurements for thermal hysteresis ΔT as a function of the number of iterations of the loop of Fig. 1 compared with the predictions (inset). Iteration 0 is the original training set of 22 alloys. At each iteration (from 1 onwards), four new predicted alloys are synthesized. The difference between the predicted and measured values of ΔT is large for iterations 1 and 2, drops significantly for iterations 3–6 and then increases beyond iteration 7. We interpret this as illustrating exploration in the early iterations, finding a reasonable minimum in the middle iterations and then exploring new areas in later iterations. (b) The ΔT as a function of the VEN feature shows that the exploration after iteration 3 is confined to an apparent minimum in the narrow interval (6.9:7; inset), favouring the B2→R transformation that is known to have the smallest ΔT (global minimum) compared with B19 and B19' transformations. (c) The average valence electron number of the four synthesized alloys as a function of the number of iterations, showing the exploratory nature (large standard deviation (s.d.)/error bars during iterations 1–2 and from 7 onwards) of the adaptive design in this feature space. The error bars denote standard deviations for VEN over the four samples. The tenth iteration indicates that the design is drifting away from the apparent global minimum (∼6.96 in the y axis).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4835535&req=5

f3: Results and insights from inference and global optimization.(a) The experimental measurements for thermal hysteresis ΔT as a function of the number of iterations of the loop of Fig. 1 compared with the predictions (inset). Iteration 0 is the original training set of 22 alloys. At each iteration (from 1 onwards), four new predicted alloys are synthesized. The difference between the predicted and measured values of ΔT is large for iterations 1 and 2, drops significantly for iterations 3–6 and then increases beyond iteration 7. We interpret this as illustrating exploration in the early iterations, finding a reasonable minimum in the middle iterations and then exploring new areas in later iterations. (b) The ΔT as a function of the VEN feature shows that the exploration after iteration 3 is confined to an apparent minimum in the narrow interval (6.9:7; inset), favouring the B2→R transformation that is known to have the smallest ΔT (global minimum) compared with B19 and B19' transformations. (c) The average valence electron number of the four synthesized alloys as a function of the number of iterations, showing the exploratory nature (large standard deviation (s.d.)/error bars during iterations 1–2 and from 7 onwards) of the adaptive design in this feature space. The error bars denote standard deviations for VEN over the four samples. The tenth iteration indicates that the design is drifting away from the apparent global minimum (∼6.96 in the y axis).
Mentions: In total, we performed nine design iterations and the results are shown in Fig. 3a, which depicts how the experimental (as well as predicted, inset) ΔT behaves with successive iterations. The range (max–min) in ΔT is large in the first two iterations, becomes smaller in iterations 3–6 and increases from the seventh iteration onwards. Figure 3b shows how the measured ΔT varies with the average VEN (one of the features used in the inference step). There are two local minima in the VEN landscape, one at 6.95 and the other at 7.13, and SVRrbf:KG predominantly explores the minima for the VEN at ∼6.95, which eventually led to the discovery of the Ti50.0Ni46.7Cu0.8Fe2.3Pd0.2 alloy. It can be seen in Fig. 3a,b that from the third iteration onwards, our strategy produces results in the desired direction of minimizing ΔT. However, as shown in Fig. 3c, after the sixth iteration our design drifts away from the apparent global minimum. We found 14 new alloys, out of 36 synthesized compositions from 9 feedback loops (see Supplementary Table 2), with ΔT <3.15 K (the best value in our original training set). One of the alloys, Ti50.0Ni46.7Cu0.8Fe2.3Pd0.2 with B2 to R transformation, discovered in iteration 6 had a ΔT of 1.84 K (as measured from DSC curves), surpassing the best value in the training data by 42%. In Table 1, we list the best five low ΔT alloys from this work.

Bottom Line: Here we show how an adaptive design strategy, tightly coupled with experiments, can accelerate the discovery process by sequentially identifying the next experiments or calculations, to effectively navigate the complex search space.We synthesize and characterize 36 predicted compositions (9 feedback loops) from a potential space of ∼800,000 compositions.Of these, 14 had smaller ΔT than any of the 22 in the original data set.

View Article: PubMed Central - PubMed

Affiliation: Theoretical Division, Los Alamos National Laboratory, MS-B262, Los Alamos, New Mexico 87545, USA.

ABSTRACT
Finding new materials with targeted properties has traditionally been guided by intuition, and trial and error. With increasing chemical complexity, the combinatorial possibilities are too large for an Edisonian approach to be practical. Here we show how an adaptive design strategy, tightly coupled with experiments, can accelerate the discovery process by sequentially identifying the next experiments or calculations, to effectively navigate the complex search space. Our strategy uses inference and global optimization to balance the trade-off between exploitation and exploration of the search space. We demonstrate this by finding very low thermal hysteresis (ΔT) NiTi-based shape memory alloys, with Ti50.0Ni46.7Cu0.8Fe2.3Pd0.2 possessing the smallest ΔT (1.84 K). We synthesize and characterize 36 predicted compositions (9 feedback loops) from a potential space of ∼800,000 compositions. Of these, 14 had smaller ΔT than any of the 22 in the original data set.

No MeSH data available.


Related in: MedlinePlus