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

Experimental measurements for the predicted Ti50.0Ni46.7Cu0.8Fe2.3Pd0.2 alloy.(a) Resistivity measurements for the new alloy, Ti50.0Ni46.7Cu0.8Fe2.3Pd0.2 , compared with NiTi (inset) emphasize the very small hysteresis (0.84 K). (b) DSC curves for Ti50.0Ni46.7Cu0.8Fe2.3Pd0.2 , whose peak-to-peak ΔT is measured as 1.84 K, which is the lowest among related NiTi-based SMAs. Thermal cycles (60 heating and cooling cycles) also show very small shift (∼0.02 K in the inset), indicating excellent thermal fatigue resistance.
© Copyright Policy - open-access
Related In: Results  -  Collection

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

f4: Experimental measurements for the predicted Ti50.0Ni46.7Cu0.8Fe2.3Pd0.2 alloy.(a) Resistivity measurements for the new alloy, Ti50.0Ni46.7Cu0.8Fe2.3Pd0.2 , compared with NiTi (inset) emphasize the very small hysteresis (0.84 K). (b) DSC curves for Ti50.0Ni46.7Cu0.8Fe2.3Pd0.2 , whose peak-to-peak ΔT is measured as 1.84 K, which is the lowest among related NiTi-based SMAs. Thermal cycles (60 heating and cooling cycles) also show very small shift (∼0.02 K in the inset), indicating excellent thermal fatigue resistance.

Mentions: It is interesting to find that from the seventh iteration onwards, the spread in ΔT (see Fig. 3a) begins to widen, relative to earlier iterations. This trend could be misconstrued as arising from model overfitting. We interpret this behaviour as a consequence of our global optimization. Recall that the purpose of SVRrbf:KG is to balance the trade-off between exploration and exploitation. As a result, every new set of experiments are purposefully designed to rapidly learn the response surface in the high-dimensional space and minimize the model uncertainties. Therefore, Fig. 3a–c indicates that the SVRrbf:KG has explored the minimum in the vicinity of VEN∼6.95 and now it is moving away in search of other local minima in our response surface. We partially capture this trend in Fig. 3c, where the trend in VEN values increases continuously (it is partial, because we are in a six-dimensional feature space). Figure 4a compares the resistivity versus temperature curves (R(T)) of Ni50Ti50 and our newly found Ti50.0Ni46.7Cu0.8Fe2.3Pd0.2. Both R(T) curves show a reversible martensitic phase transformation but our alloy also possesses negligibly small hysteresis of 0.84 K from R(T). This thermal hysteresis is consistent with a small ΔT of 1.84 K measured from DSC (Fig. 4b) and the shift in transformation temperature is negligibly small, that is, ∼0.02 K (inset in Fig. 4b). In the literature, TiNiCuPd has been reported with ‘near-zero' thermal hysteresis from resistivity measurements10. However, for the same alloy if we use our ΔT yardstick, then it is determined to be 16 K. We have listed in the Supplementary Table 2 the transformation types for all the alloys we synthesized by our design loop. There are several among these, which undergo the B2 to B19 transformation. Our best B2 to B19 alloy from the design loop has a thermal hysteresis of 9.1 K as compared with 16 K for the TiNiCuPd compound10.


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)

Experimental measurements for the predicted Ti50.0Ni46.7Cu0.8Fe2.3Pd0.2 alloy.(a) Resistivity measurements for the new alloy, Ti50.0Ni46.7Cu0.8Fe2.3Pd0.2 , compared with NiTi (inset) emphasize the very small hysteresis (0.84 K). (b) DSC curves for Ti50.0Ni46.7Cu0.8Fe2.3Pd0.2 , whose peak-to-peak ΔT is measured as 1.84 K, which is the lowest among related NiTi-based SMAs. Thermal cycles (60 heating and cooling cycles) also show very small shift (∼0.02 K in the inset), indicating excellent thermal fatigue resistance.
© Copyright Policy - open-access
Related In: Results  -  Collection

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

f4: Experimental measurements for the predicted Ti50.0Ni46.7Cu0.8Fe2.3Pd0.2 alloy.(a) Resistivity measurements for the new alloy, Ti50.0Ni46.7Cu0.8Fe2.3Pd0.2 , compared with NiTi (inset) emphasize the very small hysteresis (0.84 K). (b) DSC curves for Ti50.0Ni46.7Cu0.8Fe2.3Pd0.2 , whose peak-to-peak ΔT is measured as 1.84 K, which is the lowest among related NiTi-based SMAs. Thermal cycles (60 heating and cooling cycles) also show very small shift (∼0.02 K in the inset), indicating excellent thermal fatigue resistance.
Mentions: It is interesting to find that from the seventh iteration onwards, the spread in ΔT (see Fig. 3a) begins to widen, relative to earlier iterations. This trend could be misconstrued as arising from model overfitting. We interpret this behaviour as a consequence of our global optimization. Recall that the purpose of SVRrbf:KG is to balance the trade-off between exploration and exploitation. As a result, every new set of experiments are purposefully designed to rapidly learn the response surface in the high-dimensional space and minimize the model uncertainties. Therefore, Fig. 3a–c indicates that the SVRrbf:KG has explored the minimum in the vicinity of VEN∼6.95 and now it is moving away in search of other local minima in our response surface. We partially capture this trend in Fig. 3c, where the trend in VEN values increases continuously (it is partial, because we are in a six-dimensional feature space). Figure 4a compares the resistivity versus temperature curves (R(T)) of Ni50Ti50 and our newly found Ti50.0Ni46.7Cu0.8Fe2.3Pd0.2. Both R(T) curves show a reversible martensitic phase transformation but our alloy also possesses negligibly small hysteresis of 0.84 K from R(T). This thermal hysteresis is consistent with a small ΔT of 1.84 K measured from DSC (Fig. 4b) and the shift in transformation temperature is negligibly small, that is, ∼0.02 K (inset in Fig. 4b). In the literature, TiNiCuPd has been reported with ‘near-zero' thermal hysteresis from resistivity measurements10. However, for the same alloy if we use our ΔT yardstick, then it is determined to be 16 K. We have listed in the Supplementary Table 2 the transformation types for all the alloys we synthesized by our design loop. There are several among these, which undergo the B2 to B19 transformation. Our best B2 to B19 alloy from the design loop has a thermal hysteresis of 9.1 K as compared with 16 K for the TiNiCuPd compound10.

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