Showing 1–50 of 71 results for author: Drozdov, A

Observations of Galaxies at $z\gtrsim10$ Allow to Test Cosmological Models with Features in the Initial Power Spectrum

Authors: S. V. Pilipenko, S. A. Drozdov, M. V. Tkachev, A. G. Doroshkevich

Abstract: The initial power spectrum of density perturbations, generated during the inflationary epoch, is now constrained by observations on scales $λ>5$~Mpc and has a power-law form. The peculiarities of the inflationary process can lead to the appearance of non-power-law contributions to this spectrum, such as peaks. The exact size and shape of the peak cannot be predicted in advance. In this paper, we p… ▽ More The initial power spectrum of density perturbations, generated during the inflationary epoch, is now constrained by observations on scales $λ>5$~Mpc and has a power-law form. The peculiarities of the inflationary process can lead to the appearance of non-power-law contributions to this spectrum, such as peaks. The exact size and shape of the peak cannot be predicted in advance. In this paper, we propose methods for searching for such peaks in the region of the spectrum with $λ<5$~Mpc. Perturbations on these scales enter the nonlinear stage at $z\gtrsim10$, which is now becoming accessible to observations. Our studies of numerical models of large-scale structure with peaks in the initial spectrum have shown that spectral features on scales with $λ>0.1$~Mpc manifest in the clustering of galaxies, as well as affect their mass function, sizes, and density. Studying these characteristics of distant galaxies will allow us to constrain cosmological models with peaks. △ Less

Submitted 27 April, 2024; originally announced April 2024.

Comments: 11 pages, 5 figures, submitted to Astrophysical Bulletin

Lifetime predictions for virgin and recycled high-density polyethylene under creep conditions

Authors: A. D. Drozdov, R. Hoj Jermiin, J. deClaville Christiansen

Abstract: Recycling has become a predominant subject in industry and science due to a rising concern for the environment driven by high production volume of plastics. Replacement of virgin polymers with their recycled analogs is not always possible because recycled polymers cannot met the same property profiles as their virgin counterparts. To avoid deterioration of the mechanical properties, it is proposed… ▽ More Recycling has become a predominant subject in industry and science due to a rising concern for the environment driven by high production volume of plastics. Replacement of virgin polymers with their recycled analogs is not always possible because recycled polymers cannot met the same property profiles as their virgin counterparts. To avoid deterioration of the mechanical properties, it is proposed to replace a virgin polymer with a recycled polymer of another grade whose characteristics (measured in tensile tests) are close to those of the virgin material. This approach opens a way for the use of recycled polymers in short-term application, but its suitability for long-term applications has not yet been assessed. A thorough experimental investigation is conducted of the mechanical response of virgin high-density polyethylene (HDPE) used for insulation of pipes and recycled HDPE manufactured from post-consumer plastic waste (their stiffness, strength and elongation to break adopt similar values). A model is presented in viscoelastoplasticity of semicrystalline polymers. Its parameters are determined by matching experimental data in short-term relaxation and creep tests. The lifetime of virgin and recycled HDPE under creep conditions is evaluated by means of numerical simulation. It is shown that the stress-time to failure diagrams for virgin and recycled HDPE practically coincide. △ Less

Submitted 16 April, 2024; originally announced April 2024.

Comments: 21 pages, 10 figures

Multistage Collaborative Knowledge Distillation from a Large Language Model for Semi-Supervised Sequence Generation

Authors: Jiachen Zhao, Wenlong Zhao, Andrew Drozdov, Benjamin Rozonoyer, Md Arafat Sultan, Jay-Yoon Lee, Mohit Iyyer, Andrew McCallum

Abstract: We study semi-supervised sequence generation tasks, where the few labeled examples are too scarce to finetune a model, and meanwhile, few-shot prompted large language models (LLMs) exhibit room for improvement. In this paper, we present the discovery that a student model distilled from a few-shot prompted LLM can commonly generalize better than its teacher to unseen examples on such tasks. We find… ▽ More We study semi-supervised sequence generation tasks, where the few labeled examples are too scarce to finetune a model, and meanwhile, few-shot prompted large language models (LLMs) exhibit room for improvement. In this paper, we present the discovery that a student model distilled from a few-shot prompted LLM can commonly generalize better than its teacher to unseen examples on such tasks. We find that the student is able to learn a general pattern from the high-quality pseudolabels produced by the teacher during knowledge distillation (KD), and favorably not a general pattern from the low-quality pseudolables. Leveraging this discovery, we propose a new method, Multistage Collaborative Knowledge Distillation from an LLM (MCKD), for these tasks. MCKD first few-shot prompts an LLM to produce pseudolabels for unlabeled data. Then at each stage of an iterative KD process, a new pair of students is trained on disjoint partitions of the pseudolabeled data, and produces new and improved pseudolabels for their unseen partitions. We conduct extensive experiments on four syntactic and semantic parsing datasets and show the effectiveness of MCKD for low-resource semi-supervised sequence generation. On CRAFT biomedical parsing, for example, 3-stage MCKD with 50 labeled examples outperforms an LLM teacher and vanilla KD by 7.5% and 3.7% parsing F1, respectively, and matches the performance of supervised finetuning with 500 labeled examples. △ Less

Submitted 26 February, 2024; v1 submitted 14 November, 2023; originally announced November 2023.

PaRaDe: Passage Ranking using Demonstrations with Large Language Models

Authors: Andrew Drozdov, Honglei Zhuang, Zhuyun Dai, Zhen Qin, Razieh Rahimi, Xuanhui Wang, Dana Alon, Mohit Iyyer, Andrew McCallum, Donald Metzler, Kai Hui

Abstract: Recent studies show that large language models (LLMs) can be instructed to effectively perform zero-shot passage re-ranking, in which the results of a first stage retrieval method, such as BM25, are rated and reordered to improve relevance. In this work, we improve LLM-based re-ranking by algorithmically selecting few-shot demonstrations to include in the prompt. Our analysis investigates the cond… ▽ More Recent studies show that large language models (LLMs) can be instructed to effectively perform zero-shot passage re-ranking, in which the results of a first stage retrieval method, such as BM25, are rated and reordered to improve relevance. In this work, we improve LLM-based re-ranking by algorithmically selecting few-shot demonstrations to include in the prompt. Our analysis investigates the conditions where demonstrations are most helpful, and shows that adding even one demonstration is significantly beneficial. We propose a novel demonstration selection strategy based on difficulty rather than the commonly used semantic similarity. Furthermore, we find that demonstrations helpful for ranking are also effective at question generation. We hope our work will spur more principled research into question generation and passage ranking. △ Less

Submitted 22 October, 2023; originally announced October 2023.

Comments: Findings of EMNLP 2023

The case for studying other planetary magnetospheres and atmospheres in Heliophysics

Authors: Ian J. Cohen, Chris Arridge, Abigail Azari, Chris Bard, George Clark, Frank Crary, Shannon Curry, Peter Delamere, Ryan M. Dewey, Gina A. DiBraccio, Chuanfei Dong, Alexander Drozdov, Austin Egert, Rachael Filwett, Jasper Halekas, Alexa Halford, Andréa Hughes, Katherine Garcia-Sage, Matina Gkioulidou, Charlotte Goetz, Cesare Grava, Michael Hirsch, Hans Leo F. Huybrighs, Peter Kollmann, Laurent Lamy , et al. (15 additional authors not shown)

Abstract: Heliophysics is the field that "studies the nature of the Sun, and how it influences the very nature of space - and, in turn, the atmospheres of planetary bodies and the technology that exists there." However, NASA's Heliophysics Division tends to limit study of planetary magnetospheres and atmospheres to only those of Earth. This leaves exploration and understanding of space plasma physics at oth… ▽ More Heliophysics is the field that "studies the nature of the Sun, and how it influences the very nature of space - and, in turn, the atmospheres of planetary bodies and the technology that exists there." However, NASA's Heliophysics Division tends to limit study of planetary magnetospheres and atmospheres to only those of Earth. This leaves exploration and understanding of space plasma physics at other worlds to the purview of the Planetary Science and Astrophysics Divisions. This is detrimental to the study of space plasma physics in general since, although some cross-divisional funding opportunities do exist, vital elements of space plasma physics can be best addressed by extending the expertise of Heliophysics scientists to other stellar and planetary magnetospheres. However, the diverse worlds within the solar system provide crucial environmental conditions that are not replicated at Earth but can provide deep insight into fundamental space plasma physics processes. Studying planetary systems with Heliophysics objectives, comprehensive instrumentation, and new grant opportunities for analysis and modeling would enable a novel understanding of fundamental and universal processes of space plasma physics. As such, the Heliophysics community should be prepared to consider, prioritize, and fund dedicated Heliophysics efforts to planetary targets to specifically study space physics and aeronomy objectives. △ Less

Submitted 24 August, 2023; v1 submitted 22 August, 2023; originally announced August 2023.

KNN-LM Does Not Improve Open-ended Text Generation

Authors: Shufan Wang, Yixiao Song, Andrew Drozdov, Aparna Garimella, Varun Manjunatha, Mohit Iyyer

Abstract: In this paper, we study the generation quality of interpolation-based retrieval-augmented language models (LMs). These methods, best exemplified by the KNN-LM, interpolate the LM's predicted distribution of the next word with a distribution formed from the most relevant retrievals for a given prefix. While the KNN-LM and related methods yield impressive decreases in perplexity, we discover that th… ▽ More In this paper, we study the generation quality of interpolation-based retrieval-augmented language models (LMs). These methods, best exemplified by the KNN-LM, interpolate the LM's predicted distribution of the next word with a distribution formed from the most relevant retrievals for a given prefix. While the KNN-LM and related methods yield impressive decreases in perplexity, we discover that they do not exhibit corresponding improvements in open-ended generation quality, as measured by both automatic evaluation metrics (e.g., MAUVE) and human evaluations. Digging deeper, we find that interpolating with a retrieval distribution actually increases perplexity compared to a baseline Transformer LM for the majority of tokens in the WikiText-103 test set, even though the overall perplexity is lower due to a smaller number of tokens for which perplexity dramatically decreases after interpolation. However, when decoding a long sequence at inference time, significant improvements on this smaller subset of tokens are washed out by slightly worse predictions on most tokens. Furthermore, we discover that the entropy of the retrieval distribution increases faster than that of the base LM as the generated sequence becomes longer, which indicates that retrieval is less reliable when using model-generated text as queries (i.e., is subject to exposure bias). We hope that our analysis spurs future work on improved decoding algorithms and interpolation strategies for retrieval-augmented language models. △ Less

Submitted 23 May, 2023; originally announced May 2023.

Disc-halo gas outflows driven by stellar clusters as seen in multiwavelength tracers

Authors: Evgenii O. Vasiliev, Sergey A. Drozdov, Biman B. Nath, Ralf-Jürgen Dettmar, Yuri A. Shchekinov

Abstract: We consider the dynamics of and emission from growing superbubbles in a stratified interstellar gaseous disc driven by energy release from supernovae explosions in stellar clusters with {masses $M_{cl}= 10^5-1.6\times 10^6~M_\odot$}. Supernovae are spread randomly within a sphere of $r_c=60$ pc, and inject energy episodically with a specific rate $1/130~M_\odot^{-1}$ proportional to the star forma… ▽ More We consider the dynamics of and emission from growing superbubbles in a stratified interstellar gaseous disc driven by energy release from supernovae explosions in stellar clusters with {masses $M_{cl}= 10^5-1.6\times 10^6~M_\odot$}. Supernovae are spread randomly within a sphere of $r_c=60$ pc, and inject energy episodically with a specific rate $1/130~M_\odot^{-1}$ proportional to the star formation rate (SFR) in the cluster. Models are run for several values of SFR in the range $0.01$ to $0.1~M_\odot$ yr$^{-1}$, with the corresponding average surface energy input rate $\sim 0.04-0.4$ erg cm$^{-2}$ s$^{-1}$. We find that the discrete energy injection by isolated SNe are more efficient in blowing superbubbles: asymptotically they reach heights of up to 3 to 16 kpc for $M_{cl}=10^5-1.6\times 10^5~M_\odot$, correspondingly, and stay filled with a hot and dilute plasma for at least 30 Myr. During this time they emit X-ray, H$α$ and dust infrared emission. X-ray liminosities $L_X\propto {\rm SFR}^{3/5}$ that we derive here are consistent with observations in star-forming galaxies. Even though dust particles of small sizes $a\leq 0.03~μ$m are sputtered in the interior of bubbles, larger grains still contribute considerably ensuring the bubble luminosity $L_{\rm IR}/{\rm SFR}\sim 5\times 10^7 L_\odot M_\odot^{-1} ~{\rm yr}$. It is shown that the origin of the North Polar Spur in the Milky Way can be connected with activity of a cluster with the stellar mass of $\sim 10^5~M_\odot$ and the ${\rm SFR}\sim 0.1~M_\odot$ yr$^{-1}$ some 25--30 Myr ago. Extended luminous haloes observed in edge-on galaxies (NGC 891 as an example) can be maintained by disc spread stellar clusters of smaller masses $M_\ast \simlt 10^5~M_\odot$. △ Less

Submitted 25 January, 2023; originally announced January 2023.

Comments: 13 pages, 18 figures, accepted to MNRAS

Dust-free starburst galaxies at redshifts $z>10$

Authors: Biman B. Nath, Evgenii O. Vasiliev, Sergey A. Drozdov, Yuri A. Shchekinov

Abstract: One of the most distant galaxies GN-z11 was formed when the Universe was $\le$ 400 Myr old, and it displays a burst-like star formation rate $\sim 25~M_\odot$ yr$^{-1}$ with a metallicity $Z\sim 0.2\pm 0.1Z_\odot$. It resembles $z=2-3$ galaxies (at ``cosmic noon") except for the fact that the measured reddening $E(B-V)=0.01\pm 0.01$ indicates the presence of little or no dust. This marked absence… ▽ More One of the most distant galaxies GN-z11 was formed when the Universe was $\le$ 400 Myr old, and it displays a burst-like star formation rate $\sim 25~M_\odot$ yr$^{-1}$ with a metallicity $Z\sim 0.2\pm 0.1Z_\odot$. It resembles $z=2-3$ galaxies (at ``cosmic noon") except for the fact that the measured reddening $E(B-V)=0.01\pm 0.01$ indicates the presence of little or no dust. This marked absence of dust hints towards violent dynamical events that destroy or evacuate dust along with gas out of the galaxy on a relatively short time scale and make it transparent. We apply a 3D numerical model to infer possible physical characteristics of these events. We demonstrate that the energetics of the observed star formation rate is sufficient to tear apart the dusty veil on time scales of $20-25$ Myr. This can explain the apparent lack of evolution of UV luminosity function of galaxies between and $z\ge 10$ and $z\sim 7$, by compensating for the lower galaxy masses at higher redshift by the absence of dust. We show, however, that this is a temporary phenomenon and soon after the last of the supernovae explosions have taken place, the expanding shell shrinks and obscures the galaxy on time scales of $\approx 5-8$ Myr. △ Less

Submitted 13 February, 2023; v1 submitted 22 November, 2022; originally announced November 2022.

Comments: accepted to MNRAS, 6 pages, 3 figures

You can't pick your neighbors, or can you? When and how to rely on retrieval in the $k$NN-LM

Authors: Andrew Drozdov, Shufan Wang, Razieh Rahimi, Andrew McCallum, Hamed Zamani, Mohit Iyyer

Abstract: Retrieval-enhanced language models (LMs), which condition their predictions on text retrieved from large external datastores, have recently shown significant perplexity improvements compared to standard LMs. One such approach, the $k$NN-LM, interpolates any existing LM's predictions with the output of a $k$-nearest neighbors model and requires no additional training. In this paper, we explore the… ▽ More Retrieval-enhanced language models (LMs), which condition their predictions on text retrieved from large external datastores, have recently shown significant perplexity improvements compared to standard LMs. One such approach, the $k$NN-LM, interpolates any existing LM's predictions with the output of a $k$-nearest neighbors model and requires no additional training. In this paper, we explore the importance of lexical and semantic matching in the context of items retrieved by $k$NN-LM. We find two trends: (1) the presence of large overlap** $n$-grams between the datastore and evaluation set plays an important factor in strong performance, even when the datastore is derived from the training data; and (2) the $k$NN-LM is most beneficial when retrieved items have high semantic similarity with the query. Based on our analysis, we define a new formulation of the $k$NN-LM that uses retrieval quality to assign the interpolation coefficient. We empirically measure the effectiveness of our approach on two English language modeling datasets, Wikitext-103 and PG-19. Our re-formulation of the $k$NN-LM is beneficial in both cases, and leads to nearly 4% improvement in perplexity on the Wikitext-103 test set. △ Less

Submitted 27 October, 2022; originally announced October 2022.

Compositional Semantic Parsing with Large Language Models

Authors: Andrew Drozdov, Nathanael Schärli, Ekin Akyürek, Nathan Scales, Xinying Song, Xinyun Chen, Olivier Bousquet, Denny Zhou

Abstract: Humans can reason compositionally when presented with new tasks. Previous research shows that appropriate prompting techniques enable large language models (LLMs) to solve artificial compositional generalization tasks such as SCAN. In this work, we identify additional challenges in more realistic semantic parsing tasks with larger vocabulary and refine these prompting techniques to address them. O… ▽ More Humans can reason compositionally when presented with new tasks. Previous research shows that appropriate prompting techniques enable large language models (LLMs) to solve artificial compositional generalization tasks such as SCAN. In this work, we identify additional challenges in more realistic semantic parsing tasks with larger vocabulary and refine these prompting techniques to address them. Our best method is based on least-to-most prompting: it decomposes the problem using prompting-based syntactic parsing, then uses this decomposition to select appropriate exemplars and to sequentially generate the semantic parse. This method allows us to set a new state of the art for CFQ while requiring only 1% of the training data used by traditional approaches. Due to the general nature of our approach, we expect similar efforts will lead to new results in other tasks and domains, especially for knowledge-intensive applications. △ Less

Submitted 29 September, 2022; v1 submitted 29 September, 2022; originally announced September 2022.

Comments: Fixed metadata. No other changes

Correlation between bandwidth and frequency of plasmaspheric hiss uncovered with unsupervised machine learning

Authors: Daniel Vech, David M. Malaspina, Alexander Drozdov, Anthony Saikin

Abstract: Previous statistical studies of plasmaspheric hiss investigated the averaged shape of the magnetic field power spectra at various points in the magnetosphere. However, this approach does not consider the fact that very diverse spectral shapes exist at a given L-shell and magnetic local time. Averaging the data together means that important features of the spectral shapes are lost. In this paper, w… ▽ More Previous statistical studies of plasmaspheric hiss investigated the averaged shape of the magnetic field power spectra at various points in the magnetosphere. However, this approach does not consider the fact that very diverse spectral shapes exist at a given L-shell and magnetic local time. Averaging the data together means that important features of the spectral shapes are lost. In this paper, we use an unsupervised machine learning technique to categorize plasmaspheric hiss. In contrast to the previous studies, this technique allows us to identify power spectra that have "similar" shapes and study their spatial distribution without averaging together vastly different spectral shapes. We show that strong negative correlations exist between the hiss frequency and bandwidth, which suggests that the observed patterns are consistent with in situ wave growth. △ Less

Submitted 21 July, 2022; originally announced July 2022.

Comments: Submitted to JGR

Rheology of Polydisperse non-Spherical Graphite Particles Suspended in Mineral Oil

Authors: Th. Larsen, A. L. Søbye, J. R. Royer, W. C. K. Poon, T. Larsen, S. J. Andreasen, A. D. Drozdov, J. D. C. Christiansen

Abstract: We study the role of filler concentration and microphysics on the rheology of polydisperse flake-graphite particles suspended in Newtonian mineral oil. Under steady shear, our samples exhibit shear thinning and yielding behaviour is observed for volume fractions $φ> 0.18$. Time-temperature superposition was observed using an Arrhenius-type horizontal shift factor, giving a flow activation energy t… ▽ More We study the role of filler concentration and microphysics on the rheology of polydisperse flake-graphite particles suspended in Newtonian mineral oil. Under steady shear, our samples exhibit shear thinning and yielding behaviour is observed for volume fractions $φ> 0.18$. Time-temperature superposition was observed using an Arrhenius-type horizontal shift factor, giving a flow activation energy that is dependent on the graphite volume fraction, suggesting concentration-dependent contributions to relaxation processes in the suspensions. The flow curves are fitted by a constraint-based model, indicating that the flow behaviour is controlled by frictional and adhesive contacts, with the model suggesting that the adhesive stress is temperature dependent. △ Less

Submitted 26 October, 2022; v1 submitted 6 July, 2022; originally announced July 2022.

Inducing and Using Alignments for Transition-based AMR Parsing

Authors: Andrew Drozdov, Jiawei Zhou, Radu Florian, Andrew McCallum, Tahira Naseem, Yoon Kim, Ramon Fernandez Astudillo

Abstract: Transition-based parsers for Abstract Meaning Representation (AMR) rely on node-to-word alignments. These alignments are learned separately from parser training and require a complex pipeline of rule-based components, pre-processing, and post-processing to satisfy domain-specific constraints. Parsers also train on a point-estimate of the alignment pipeline, neglecting the uncertainty due to the in… ▽ More Transition-based parsers for Abstract Meaning Representation (AMR) rely on node-to-word alignments. These alignments are learned separately from parser training and require a complex pipeline of rule-based components, pre-processing, and post-processing to satisfy domain-specific constraints. Parsers also train on a point-estimate of the alignment pipeline, neglecting the uncertainty due to the inherent ambiguity of alignment. In this work we explore two avenues for overcoming these limitations. First, we propose a neural aligner for AMR that learns node-to-word alignments without relying on complex pipelines. We subsequently explore a tighter integration of aligner and parser training by considering a distribution over oracle action sequences arising from aligner uncertainty. Empirical results show this approach leads to more accurate alignments and generalization better from the AMR2.0 to AMR3.0 corpora. We attain a new state-of-the art for gold-only trained models, matching silver-trained performance without the need for beam search on AMR3.0. △ Less

Submitted 3 May, 2022; originally announced May 2022.

Comments: Accepted at NAACL 2022

Universal diamond edge Raman scale to 0.5 terapascal: The implication to metallization of hydrogen

Authors: M. I. Eremets, V. S. Minkov, P. P. Kong, A. P. Drozdov, S. Chariton, V. B. Prakapenka

Abstract: The recent progress in generating static pressures up to terapascal values opens opportunities for studying novel materials with unusual properties, such as metallization of hydrogen and high-temperature superconductivity. However, an evaluation of pressure above ~0.3 terapascal is a challenge. We report a universal high-pressure scale up to ~0.5 terapascal, based on the shift of the Raman edge of… ▽ More The recent progress in generating static pressures up to terapascal values opens opportunities for studying novel materials with unusual properties, such as metallization of hydrogen and high-temperature superconductivity. However, an evaluation of pressure above ~0.3 terapascal is a challenge. We report a universal high-pressure scale up to ~0.5 terapascal, based on the shift of the Raman edge of stressed diamond anvils correlated with the equation of state of Au and does not require an additional pressure sensor. According to the new scale, the pressure values are substantially lower by 20% at ~0.5 terapascal compared to the extrapolation of the existing scales. We compared the available data of H2 at the highest static pressures. We showed that the onset of the proposed metallization of molecular hydrogen reported by different groups is consistent when corrected with the new scale and can be compared with various theoretical predictions. △ Less

Submitted 14 February, 2022; originally announced February 2022.

High-temperature superconductivity in hydrides: experimental evidence and details

Authors: M. I. Eremets, V. S. Minkov, A. P. Drozdov, P. P. Kong, V. Ksenofontov, S. I. Shylin, S. L. Bud ko, R. Prozorov, F. F. Balakirev, Dan Sun, S. Mozaffari, L. Balicas

Abstract: Since the discovery of superconductivity at 200 K in H3S [1] similar or higher transition temperatures, Tcs, have been reported for various hydrogen-rich compounds under ultra-high pressures [2]. Superconductivity was experimentally proved by different methods, including electrical resistance, magnetic susceptibility, optical infrared, and nuclear resonant scattering measurements. The crystal stru… ▽ More Since the discovery of superconductivity at 200 K in H3S [1] similar or higher transition temperatures, Tcs, have been reported for various hydrogen-rich compounds under ultra-high pressures [2]. Superconductivity was experimentally proved by different methods, including electrical resistance, magnetic susceptibility, optical infrared, and nuclear resonant scattering measurements. The crystal structures of superconducting phases were determined by X-ray diffraction. Numerous electrical transport measurements demonstrate the typical behaviour of a conventional phonon-mediated superconductor: zero resistance below Tc, the shift of Tc to lower temperatures under external magnetic fields, and pronounced isotope effect. Remarkably, the results are in good agreement with the theoretical predictions, which describe superconductivity in hydrides within the framework of the conventional BCS theory. However, despite this acknowledgment, experimental evidence for the superconducting state in these compounds has recently been treated with criticism [3, 4], which apparently stems from misunderstanding and misinterpretation of complicated experiments performed under very high pressures. Here, we describe in greater detail the experiments revealing high-temperature superconductivity in hydrides under high pressures. We show that the arguments against superconductivity [3, 4] can be either refuted or explained. The experiments on the high-temperature superconductivity in hydrides clearly contradict the theory of hole superconductivity [4] and eliminate it [3]. △ Less

Submitted 13 January, 2022; originally announced January 2022.

Raman and infrared studies of CdSe/CdS core/shell nanoplatelets

Authors: Alexander I. Lebedev, Bedil M. Saidzhonov, Konstantin A. Drozdov, Andrey A. Khomich, Roman B. Vasiliev

Abstract: The vibrational spectroscopy of semiconductor nanostructures can provide important information on their structure. In this work, experimental Raman and infrared spectra are compared with vibrational spectra of CdSe/CdS core/shell nanoplatelets calculated from first principles using the density functional theory. The calculations confirm the two-mode behavior of phonon spectra of nanostructures. An… ▽ More The vibrational spectroscopy of semiconductor nanostructures can provide important information on their structure. In this work, experimental Raman and infrared spectra are compared with vibrational spectra of CdSe/CdS core/shell nanoplatelets calculated from first principles using the density functional theory. The calculations confirm the two-mode behavior of phonon spectra of nanostructures. An analysis of the experimental spectra reveals the absence of modes with a high amplitude of vibrations of surface atoms, which indicates their strong dam**. Taking into account the difference in the dam** of different modes and their calculated intensities, all bands in the spectra are unambiguously identified. It is found that the frequencies of longitudinal optical modes in heterostructures are close to the frequencies of LO phonons in bulk strained constituents, whereas the frequencies of transverse modes can differ significantly from those of the corresponding TO phonons. It is shown that an anomalous thickness dependence of CdS TO mode is due to a noticeable surface relaxation of the outer Cd layer in the nanostructure. △ Less

Submitted 12 December, 2021; originally announced December 2021.

Comments: 9 pages, 9 figures, 2 tables

Journal ref: J. Phys. Chem. C 125, 6758 (2021)

Infrared imaging of samples in ultra high pressure diamond anvil cells

Authors: Tarun Patel, A. Drozdov, V. S. Minkov, M. I. Eremets, E. J. Nicol, J. P. Carbotte, T. Timusk, A. W. Tsen

Abstract: We describe an experimental platform that generates infrared images of micrometer-sized samples in the high pressure region of a diamond anvil cell. Using a 2.3 micron laser as a source of radiation, the system will be particularly useful in identifying hydride superconductors which exhibit an anomalous temperature dependence of reflectivity in the 2.3 micron region. Our system shows an intensity… ▽ More We describe an experimental platform that generates infrared images of micrometer-sized samples in the high pressure region of a diamond anvil cell. Using a 2.3 micron laser as a source of radiation, the system will be particularly useful in identifying hydride superconductors which exhibit an anomalous temperature dependence of reflectivity in the 2.3 micron region. Our system shows an intensity stability within one percent when the sample temperature is swept from 100 K to 300 K. The spatial stability is of the order of a few micrometers in the same temperature range. △ Less

Submitted 26 October, 2021; originally announced October 2021.

Metallization of hydrogen

Authors: M. I. Eremets, P. P. Kong, A. P. Drozdov

Abstract: In previous work,1 we showed that hydrogen metallizes in phase III at temperatures below ~200 K and at pressures near ~350 GPa. Here, we perform a detailed study of electrical conductivity R(T) in phase III over a pressure range of 200-400 GPa and a temperature range of 80-300 K, and we show that hydrogen transforms from a semiconducting to a metallic state already at ~315 GPa. This transformation… ▽ More In previous work,1 we showed that hydrogen metallizes in phase III at temperatures below ~200 K and at pressures near ~350 GPa. Here, we perform a detailed study of electrical conductivity R(T) in phase III over a pressure range of 200-400 GPa and a temperature range of 80-300 K, and we show that hydrogen transforms from a semiconducting to a metallic state already at ~315 GPa. This transformation is also supported by Raman spectroscopy: the Raman signal intensity decreases with pressure in accordance with the appearance and increase of electrical conductivity. Moreover, the Raman and electrical measurements yield the same boundary between the hydrogen phases III and V. △ Less

Submitted 22 September, 2021; originally announced September 2021.

Improved Latent Tree Induction with Distant Supervision via Span Constraints

Authors: Zhiyang Xu, Andrew Drozdov, Jay Yoon Lee, Tim O'Gorman, Subendhu Rongali, Dylan Finkbeiner, Shilpa Suresh, Mohit Iyyer, Andrew McCallum

Abstract: For over thirty years, researchers have developed and analyzed methods for latent tree induction as an approach for unsupervised syntactic parsing. Nonetheless, modern systems still do not perform well enough compared to their supervised counterparts to have any practical use as structural annotation of text. In this work, we present a technique that uses distant supervision in the form of span co… ▽ More For over thirty years, researchers have developed and analyzed methods for latent tree induction as an approach for unsupervised syntactic parsing. Nonetheless, modern systems still do not perform well enough compared to their supervised counterparts to have any practical use as structural annotation of text. In this work, we present a technique that uses distant supervision in the form of span constraints (i.e. phrase bracketing) to improve performance in unsupervised constituency parsing. Using a relatively small number of span constraints we can substantially improve the output from DIORA, an already competitive unsupervised parsing system. Compared with full parse tree annotation, span constraints can be acquired with minimal effort, such as with a lexicon derived from Wikipedia, to find exact text matches. Our experiments show span constraints based on entities improves constituency parsing on English WSJ Penn Treebank by more than 5 F1. Furthermore, our method extends to any domain where span constraints are easily attainable, and as a case study we demonstrate its effectiveness by parsing biomedical text from the CRAFT dataset. △ Less

Submitted 1 November, 2021; v1 submitted 10 September, 2021; originally announced September 2021.

Comments: EMNLP 2021

Thermal Transport in Graphene Composites: The Effect of Lateral Dimensions of Graphene Fillers

Authors: Sriharsha Sudhindra, Farnia Rashvand, Dylan Wright, Zahra Barani, Aleksey D. Drozdov, Saba Baraghani, Claudia Backes, Fariborz Kargar, Alexander A. Balandin

Abstract: We report on the investigation of thermal transport in non-cured silicone composites with graphene fillers of different lateral dimensions. Graphene fillers are comprised of few-layer graphene flakes with lateral sizes in the range from 400 nm to 1200 nm and number of atomic planes from one to ~100. The distribution of the lateral dimensions and thicknesses of graphene fillers has been determined… ▽ More We report on the investigation of thermal transport in non-cured silicone composites with graphene fillers of different lateral dimensions. Graphene fillers are comprised of few-layer graphene flakes with lateral sizes in the range from 400 nm to 1200 nm and number of atomic planes from one to ~100. The distribution of the lateral dimensions and thicknesses of graphene fillers has been determined via atomic force microscopy statistics. It was found that in the examined range of the lateral dimensions the thermal conductivity of the composites increases with the increasing size of the graphene fillers. The observed difference in thermal properties can be related to the average gray phonon mean free path in graphene, which has been estimated to be around ~800 nm at room temperature. The thermal contact resistance of composites with graphene fillers of 1200-nm lateral dimensions was also smaller than that of composites with graphene fillers of 400-nm lateral dimensions. The effects of the filler loading fraction and the filler size on the thermal conductivity of the composites were rationalized within the Kanari model. The obtained results are important for optimization of graphene fillers for applications in thermal interface materials for heat removal from high-power-density electronics. △ Less

Submitted 18 August, 2021; originally announced August 2021.

Comments: 38 pages, 6 figures

Machine-learning based discovery of missing physical processes in radiation belt modeling

Authors: Enrico Camporeale, George J. Wilkie, Alexander Drozdov, Jacob Bortnik

Abstract: Real-time prediction of the dynamics of energetic electrons in Earth's radiation belts incorporating incomplete observation data is important to protect valuable artificial satellites and to understand their physical processes. Traditionally, reduced models have employed a diffusion equation based on the quasilinear approximation. Using a Physics-Informed Neural Network (PINN) framework, we train… ▽ More Real-time prediction of the dynamics of energetic electrons in Earth's radiation belts incorporating incomplete observation data is important to protect valuable artificial satellites and to understand their physical processes. Traditionally, reduced models have employed a diffusion equation based on the quasilinear approximation. Using a Physics-Informed Neural Network (PINN) framework, we train and test a model based on Van Allen Probe data. We present a recipe for gleaning physical insight from solving the ill-posed inverse problem of inferring model coefficients from data using PINNs. With this, it is discovered that the dynamics of "killer electrons" is described more accurately instead by a drift-diffusion equation. A parameterization for the diffusion and drift coefficients, which is both simpler and more accurate than existing models, is presented. △ Less

Submitted 12 November, 2021; v1 submitted 29 July, 2021; originally announced July 2021.

Comments: 9 pages, 8 figures, under review

Measurement of the proton spin structure at long distances

Authors: X. Zheng, A. Deur, H. Kang, S. E. Kuhn, M. Ripani, J. Zhang, K. P. Adhikari, S. Adhikari, M. J. Amaryan, H. Atac, H. Avakian, L. Barion, M. Battaglieri, I. Bedlinskiy, F. Benmokhtar, A. Bianconi, A. S. Biselli, S. Boiarinov, M. Bondi, F. Bossu, P. Bosted, W. J. Briscoe, J. Brock, W. K. Brooks, D. Bulumulla , et al. (126 additional authors not shown)

Abstract: Measuring the spin structure of protons and neutrons tests our understanding of how they arise from quarks and gluons, the fundamental building blocks of nuclear matter. At long distances the coupling constant of the strong interaction becomes large, requiring non-perturbative methods to calculate quantum chromodynamics processes, such as lattice gauge theory or effective field theories. Here we r… ▽ More Measuring the spin structure of protons and neutrons tests our understanding of how they arise from quarks and gluons, the fundamental building blocks of nuclear matter. At long distances the coupling constant of the strong interaction becomes large, requiring non-perturbative methods to calculate quantum chromodynamics processes, such as lattice gauge theory or effective field theories. Here we report proton spin structure measurements from scattering a polarized electron beam off polarized protons. The spin-dependent cross-sections were measured at large distances, corresponding to the region of low momentum transfer squared between 0.012 and 1.0 GeV$^2$. This kinematic range provides unique tests of chiral effective field theory predictions. Our results show that a complete description of the nucleon spin remains elusive, and call for further theoretical works, e.g. in lattice quantum chromodynamics. Finally, our data extrapolated to the photon point agree with the Gerasimov-Drell-Hearn sum rule, a fundamental prediction of quantum field theory that relates the anomalous magnetic moment of the proton to its integrated spin-dependent cross-sections. △ Less

Submitted 12 January, 2022; v1 submitted 4 February, 2021; originally announced February 2021.

Comments: Published version. 10 pages, 5 figures. 20 pages of supplementary material (data tables and a figure)

Report number: JLAB-PHY-20-3251, DOE/OR/23177-5042

Journal ref: Nature Physics, vo. 17 736-741 (2021)

Bayesian inference of quasi-linear radial diffusion parameters using Van Allen Probes

Authors: Rakesh Sarma, Mandar Chandorkar, Irina Zhelavskaya, Yuri Shprits, Alexander Drozdov, Enrico Camporeale

Abstract: The Van Allen radiation belts in the magnetosphere have been extensively studied using models based on radial diffusion theory, which is based on a quasi-linear approach with prescribed inner and outer boundary conditions. The 1-d diffusion model requires the knowledge of a diffusion coefficient and an electron loss timescale, which are typically parameterized in terms of various quantities such a… ▽ More The Van Allen radiation belts in the magnetosphere have been extensively studied using models based on radial diffusion theory, which is based on a quasi-linear approach with prescribed inner and outer boundary conditions. The 1-d diffusion model requires the knowledge of a diffusion coefficient and an electron loss timescale, which are typically parameterized in terms of various quantities such as the spatial ($L$) coordinate or a geomagnetic index (for example, $Kp$). These terms are empirically derived, not directly measurable, and hence are not known precisely, due to the inherent non-linearity of the process and the variable boundary conditions. In this work, we demonstrate a probabilistic approach by inferring the values of the diffusion and loss term parameters, along with their uncertainty, in a Bayesian framework, where identification is obtained using the Van Allen Probe measurements. Our results show that the probabilistic approach statistically improves the performance of the model, compared to the parameterization employed in the literature. △ Less

Submitted 7 February, 2020; originally announced February 2020.

Superconductivity up to 243 K in yttrium hydrides under high pressure

Authors: P. P. Kong, V. S. Minkov, M. A. Kuzovnikov, S. P. Besedin, A. P. Drozdov, S. Mozaffari, L. Balicas, F. F. Balakirev, V. B. Prakapenka, E. Greenberg, D. A. Knyazev, M. I. Eremets

Abstract: The discovery of high-temperature conventional superconductivity in H3S with a critical temperature of Tc=203 K was followed by the recent record of Tc ~250 K in the face-centered cubic (fcc) lanthanum hydride LaH10 compound. It was realized in a new class of hydrogen-dominated compounds having a clathrate-like crystal structure in which hydrogen atoms form a 3D framework and surround a host atom… ▽ More The discovery of high-temperature conventional superconductivity in H3S with a critical temperature of Tc=203 K was followed by the recent record of Tc ~250 K in the face-centered cubic (fcc) lanthanum hydride LaH10 compound. It was realized in a new class of hydrogen-dominated compounds having a clathrate-like crystal structure in which hydrogen atoms form a 3D framework and surround a host atom of rare earth elements. Yttrium hydrides are predicted to have even higher Tc exceeding room temperature. In this paper, we synthesized and refined the crystal structure of new hydrides: YH4, YH6, and YH9 at pressures up to 237 GPa finding that YH4 crystalizes in the I4/mmm lattice, YH6 in Im-3m lattice and YH9 in P63/mmc lattice in excellent agreement with the calculations. The observed very high-temperature superconductivity is comparable to that found in fcc-LaH10: the pressure dependence of Tc for YH9 also displays a "dome like shape" with the highest Tc of 243 K at 201 GPa. We also observed a Tc of 227 K at 237 GPa for the YH6 phase. However, the measured Tcs are notably lower by ~30 K than predicted. Evidence for superconductivity includes the observation of zero electrical resistance, a decrease of Tc under an external magnetic field and an isotope effect. The theoretically predicted fcc YH10 with the promising highest Tc>300 K was not stabilized in our experiments under pressures up to 237 GPa. △ Less

Submitted 23 September, 2019; originally announced September 2019.

Simulations of the inner magnetospheric energetic electrons using theIMPTAM-VERB coupled model

Authors: Angelica M. Castillo, Yuri Y. Shprits, Natalia Ganushkina, Alexander Drozdov, Nikita Aseev, Dedong Wang, Stepan Dubyagin

Abstract: In this study, we present initial results of the coupling between the Inner Magnetospheric Particle Transport and Acceleration Model (IMPTAM) and the Versatile Electron Radiation Belt (VERB-3D) code. IMPTAM traces electrons of 10 - 100 keV energies from the plasma sheet (L = 9 Re) to inner L-shell regions. The flux evolution modeled by IMPTAM is used at the low energy and outer L* computational bo… ▽ More In this study, we present initial results of the coupling between the Inner Magnetospheric Particle Transport and Acceleration Model (IMPTAM) and the Versatile Electron Radiation Belt (VERB-3D) code. IMPTAM traces electrons of 10 - 100 keV energies from the plasma sheet (L = 9 Re) to inner L-shell regions. The flux evolution modeled by IMPTAM is used at the low energy and outer L* computational boundaries of the VERB code (assuming a dipole approximation) to perform radiation belt simulations of energetic electrons. The model was tested on the March 17th, 2013 storm, for a six-day period. Four different simulations were performed and their results compared to satellites observations from Van Allen probes and GOES. The coupled IMPTAM-VERB model reproduces evolution and storm-time features of electron fluxes throughout the studied storm in agreement with the satellite data (within ~0.5 orders of magnitude). Including dynamics of the low energy population at L* = 6.6 increases fluxes closer to the heart of the belt and has a strong impact in the VERB simulations at all energies. However, inclusion of magnetopause losses leads to drastic flux decreases even below L* = 3. The dynamics of low energy electrons (max. 10s of keV) do not affect electron fluxes at energies >= 900 keV. Since the IMPTAM-VERB coupled model is only driven by solar wind parameters and the Dst and Kp indexes, it is suitable as a forecasting tool. In this study, we demonstrate that the estimation of electron dynamics with satellite-data-independent models is possible and very accurate. △ Less

Submitted 18 September, 2019; originally announced September 2019.

Unsupervised Latent Tree Induction with Deep Inside-Outside Recursive Autoencoders

Authors: Andrew Drozdov, Pat Verga, Mohit Yadav, Mohit Iyyer, Andrew McCallum

Abstract: We introduce deep inside-outside recursive autoencoders (DIORA), a fully-unsupervised method for discovering syntax that simultaneously learns representations for constituents within the induced tree. Our approach predicts each word in an input sentence conditioned on the rest of the sentence and uses inside-outside dynamic programming to consider all possible binary trees over the sentence. At te… ▽ More We introduce deep inside-outside recursive autoencoders (DIORA), a fully-unsupervised method for discovering syntax that simultaneously learns representations for constituents within the induced tree. Our approach predicts each word in an input sentence conditioned on the rest of the sentence and uses inside-outside dynamic programming to consider all possible binary trees over the sentence. At test time the CKY algorithm extracts the highest scoring parse. DIORA achieves a new state-of-the-art F1 in unsupervised binary constituency parsing (unlabeled) in two benchmark datasets, WSJ and MultiNLI. △ Less

Submitted 4 April, 2019; v1 submitted 3 April, 2019; originally announced April 2019.

Comments: 14 pages, 8 figures, 8 tables. NAACL 2019

Superconductivity at 250 K in lanthanum hydride under high pressures

Authors: A. P. Drozdov, P. P. Kong, V. S. Minkov, S. P. Besedin, M. A. Kuzovnikov, S. Mozaffari, L. Balicas, F. Balakirev, D. Graf, V. B. Prakapenka, E. Greenberg, D. A. Knyazev, M. Tkacz, M. I. Eremets

Abstract: The discovery of superconductivity at 203 K in H3S brought attention back to conventional superconductors whose properties can be described by the Bardeen-Cooper-Schrieffer (BCS) and the Migdal-Eliashberg theories. These theories predict that high, and even room temperature superconductivity (RTSC) is possible in metals possessing certain favorable parameters such as lattice vibrations at high fre… ▽ More The discovery of superconductivity at 203 K in H3S brought attention back to conventional superconductors whose properties can be described by the Bardeen-Cooper-Schrieffer (BCS) and the Migdal-Eliashberg theories. These theories predict that high, and even room temperature superconductivity (RTSC) is possible in metals possessing certain favorable parameters such as lattice vibrations at high frequencies. However, these general theories do not suffice to predict real superconductors. New superconducting materials can be predicted now with the aid of first principles calculations based on Density Functional Theory (DFT). In particular, the calculations suggested a new family of hydrides possessing a clathrate structure, where the host atom (Ca, Y, La) is at the center of the cage formed by hydrogen atoms. For LaH10 and YH10 superconductivity, with critical temperatures Tc ranging between 240 and 320 K is predicted at megabar pressures. Here, we report superconductivity with a record Tc ~ 250 K within the Fm-3m structure of LaH10 at a pressure P ~ 170 GPa. We proved the existence of superconductivity at 250 K through the observation of zero-resistance, isotope effect, and the decrease of Tc under an external magnetic field, which suggests an upper critical magnetic field of 120 T at zero-temperature. The pressure dependence of the transition temperatures Tc (P) has a maximum of 250-252 K at the pressure of about 170 GPa. This leap, by ~ 50 K, from the previous Tc record of 203 K indicates the real possibility of achieving RTSC (that is at 273 K) in the near future at high pressures and the perspective of conventional superconductivity at ambient pressure. △ Less

Submitted 4 December, 2018; originally announced December 2018.

Superconductivity at 215 K in lanthanum hydride at high pressures

Authors: A. P. Drozdov, V. S. Minkov, S. P. Besedin, P. P. Kong, M. A. Kuzovnikov, D. A. Knyazev, M. I. Eremets

Abstract: We synthesized lanthanum hydride (LaHx) by laser heating of lanthanum in hydrogen atmosphere at pressure P = 170 GPa. The sample shows a superconducting step at 209 K and 170 GPa and temperature dependence of resistance. By releasing the pressure to 150 GPa, the superconducting transition temperature Tc increases to 215 K - the record Tc. This finding supports a way of achieving Tc higher than the… ▽ More We synthesized lanthanum hydride (LaHx) by laser heating of lanthanum in hydrogen atmosphere at pressure P = 170 GPa. The sample shows a superconducting step at 209 K and 170 GPa and temperature dependence of resistance. By releasing the pressure to 150 GPa, the superconducting transition temperature Tc increases to 215 K - the record Tc. This finding supports a way of achieving Tc higher than the one in H3S (203 K) in hydrides with sodalite-like structures, first proposed for CaH6 (Tc=245 K) and later for yttrium and lanthanum hydrides where higher, room temperature superconductivity is expected. △ Less

Submitted 21 August, 2018; originally announced August 2018.

Do latent tree learning models identify meaningful structure in sentences?

Authors: Adina Williams, Andrew Drozdov, Samuel R. Bowman

Abstract: Recent work on the problem of latent tree learning has made it possible to train neural networks that learn to both parse a sentence and use the resulting parse to interpret the sentence, all without exposure to ground-truth parse trees at training time. Surprisingly, these models often perform better at sentence understanding tasks than models that use parse trees from conventional parsers. This… ▽ More Recent work on the problem of latent tree learning has made it possible to train neural networks that learn to both parse a sentence and use the resulting parse to interpret the sentence, all without exposure to ground-truth parse trees at training time. Surprisingly, these models often perform better at sentence understanding tasks than models that use parse trees from conventional parsers. This paper aims to investigate what these latent tree learning models learn. We replicate two such models in a shared codebase and find that (i) only one of these models outperforms conventional tree-structured models on sentence classification, (ii) its parsing strategies are not especially consistent across random restarts, (iii) the parses it produces tend to be shallower than standard Penn Treebank (PTB) parses, and (iv) they do not resemble those of PTB or any other semantic or syntactic formalism that the authors are aware of. △ Less

Submitted 26 February, 2018; v1 submitted 4 September, 2017; originally announced September 2017.

Comments: 15 pages, 6 figures, 4 tables. v1. was submitted to TACL, v2. was accepted to TACL, name change, additional baselines (R/L branching)

Semimetallic molecular hydrogen at pressure above 350 GPa

Authors: M. I. Eremets, A. P. Drozdov, P. P. Kong, H. Wang

Abstract: According to the theoretical predictions, insulating molecular hydrogen dissociates and transforms to an atomic metal at pressures P~370-500 GPa. In another scenario, the metallization first occurs in the 250-500 GPa pressure range in molecular hydrogen through overlap** of electronic bands. The calculations are not accurate enough to predict which option is realized. Here we show that at a pres… ▽ More According to the theoretical predictions, insulating molecular hydrogen dissociates and transforms to an atomic metal at pressures P~370-500 GPa. In another scenario, the metallization first occurs in the 250-500 GPa pressure range in molecular hydrogen through overlap** of electronic bands. The calculations are not accurate enough to predict which option is realized. Here we show that at a pressure of ~360 GPa and temperatures <200 K the hydrogen starts to conduct, and that temperature dependence of the electrical conductivity is typical of a semimetal. The conductivity, measured up to 440 GPa, increases strongly with pressure. Raman spectra, measured up to 480 GPa, indicate that hydrogen remains a molecular solid at pressures up to 440 GPa, while at higher pressures the Raman signal vanishes, likely indicating further transformation to a good molecular metal or to an atomic state. △ Less

Submitted 14 June, 2019; v1 submitted 17 August, 2017; originally announced August 2017.

Emergent Communication in a Multi-Modal, Multi-Step Referential Game

Authors: Katrina Evtimova, Andrew Drozdov, Douwe Kiela, Kyunghyun Cho

Abstract: Inspired by previous work on emergent communication in referential games, we propose a novel multi-modal, multi-step referential game, where the sender and receiver have access to distinct modalities of an object, and their information exchange is bidirectional and of arbitrary duration. The multi-modal multi-step setting allows agents to develop an internal communication significantly closer to n… ▽ More Inspired by previous work on emergent communication in referential games, we propose a novel multi-modal, multi-step referential game, where the sender and receiver have access to distinct modalities of an object, and their information exchange is bidirectional and of arbitrary duration. The multi-modal multi-step setting allows agents to develop an internal communication significantly closer to natural language, in that they share a single set of messages, and that the length of the conversation may vary according to the difficulty of the task. We examine these properties empirically using a dataset consisting of images and textual descriptions of mammals, where the agents are tasked with identifying the correct object. Our experiments indicate that a robust and efficient communication protocol emerges, where gradual information exchange informs better predictions and higher communication bandwidth improves generalization. △ Less

Submitted 16 April, 2018; v1 submitted 29 May, 2017; originally announced May 2017.

Comments: Published as a conference paper at ICLR 2018. 12 pages

Elemental Phosphorus: structural and superconducting phase diagram under pressure

Authors: José A. Flores-Livas, Antonio Sanna, Alexander P. Drozdov, Lilia Boeri, Gianni Profeta, Mikhail Eremets, Stefan Goedecker

Abstract: Pressure-induced superconductivity and structural phase transitions in phosphorous (P) are studied by resistivity measurements under pressures up to 170 GPa and fully $ab-initio$ crystal structure and superconductivity calculations up to 350 GPa. Two distinct superconducting transition temperature (T$_{c}$) vs. pressure ($P$) trends at low pressure have been reported more than 30 years ago, and fo… ▽ More Pressure-induced superconductivity and structural phase transitions in phosphorous (P) are studied by resistivity measurements under pressures up to 170 GPa and fully $ab-initio$ crystal structure and superconductivity calculations up to 350 GPa. Two distinct superconducting transition temperature (T$_{c}$) vs. pressure ($P$) trends at low pressure have been reported more than 30 years ago, and for the first time we are able to reproduce them and devise a consistent explanation founded on thermodynamically metastable phases of black-phosphorous. Our experimental and theoretical results form a single, consistent picture which not only provides a clear understanding of elemental P under pressure but also sheds light on the long-standing and unsolved $anomalous$ superconductivity trend. Moreover, at higher pressures we predict a similar scenario of multiple metastable structures which coexist beyond their thermodynamical stability range. Metastable phases of P experimentally accessible at pressures above 240 GPa should exhibit T$_{c}$'s as high as 15 K, i.e. three times larger than the predicted value for the ground-state crystal structure. We observe that all the metastable structures systematically exhibit larger transition temperatures than the ground-state ones, indicating that the exploration of metastable phases represents a promising route to design materials with improved superconducting properties. △ Less

Submitted 16 March, 2017; originally announced March 2017.

Comments: 14 pages, 4 figures

Journal ref: Phys. Rev. Materials 1, 024802 (2017)

Comments on the claimed observation of the Wigner-Huntington Transition to Metallic Hydrogen

Authors: M. I. Eremets, A. P. Drozdov

Abstract: In their recent work Dias and Silvera (Science 2017) claim to have observed the Wigner-Huntington transition of hydrogen to a metallic state (MH) at a pressure of 495 GPa at low temperatures. The evidence for this transition is based on a high electron carrier density deduced from a Drude free electron model fitted to the reflectivity of the sample. Based on our analysis of the reflectivity data w… ▽ More In their recent work Dias and Silvera (Science 2017) claim to have observed the Wigner-Huntington transition of hydrogen to a metallic state (MH) at a pressure of 495 GPa at low temperatures. The evidence for this transition is based on a high electron carrier density deduced from a Drude free electron model fitted to the reflectivity of the sample. Based on our analysis of the reflectivity data we find no convincing evidence for metallic hydrogen in their published data. The pressure determination is also ambiguous - it should be ~630 GPa according to the presented Raman spectrum. For comparison, we present our own data on the observation of highly reflecting hydrogen at pressures of 350-400 GPa. The appearance of metallic reflectivity is accompanied with a finite electrical conductivity of the sample. We argue that the actual pressure in the experiment of Dias and Silvera is likely below 400 GPa. In this case the observed enhanced reflectivity would be related to the phase transformation to conductive state published in arXiv:1601.04479. △ Less

Submitted 16 February, 2017; originally announced February 2017.

Spectroscopy of H$_3$S: evidence of a new energy scale for superconductivity

Authors: F. Capitani, B. Langerome, J. -B. Brubach, P. Roy, A. Drozdov, M. I. Eremets, E. J. Nicol, J. P. Carbotte, T. Timusk

Abstract: The discovery of a superconducting phase in sulfur hydride under high pressure with a critical temperature above 200 K has provided a new impetus to the search for even higher $T_c$. Theory predicted and experiment confirmed that the phase involved is H$_3$S with Im-3m crystal structure. The observation of a sharp drop in resistance to zero at $T_c$, its downward shift with magnetic field and a Me… ▽ More The discovery of a superconducting phase in sulfur hydride under high pressure with a critical temperature above 200 K has provided a new impetus to the search for even higher $T_c$. Theory predicted and experiment confirmed that the phase involved is H$_3$S with Im-3m crystal structure. The observation of a sharp drop in resistance to zero at $T_c$, its downward shift with magnetic field and a Meissner effect confirm superconductivity but the mechanism involved remains to be determined. Here, we provide a first optical spectroscopy study of this new superconductor. Experimental results for the optical reflectivity of H$_3$S, under high pressure of 150 GPa, for several temperatures and over the range 60 to 600 meV of photon energies, are compared with theoretical calculations based on Eliashberg theory using DFT results for the electron-phonon spectral density $α^2$F($Ω$). Two significant features stand out: some remarkably strong infrared active phonons at $\approx$ 160 meV and a band with a depressed reflectance in the superconducting state in the region from 450 meV to 600 meV. In this energy range, as predicted by theory, H$_3$S is found to become a better reflector with increasing temperature. This temperature evolution is traced to superconductivity originating from the electron-phonon interaction. The shape, magnitude, and energy dependence of this band at 150 K agrees with our calculations. This provides strong evidence of a conventional mechanism. However, the unusually strong optical phonon suggests a contribution of electronic degrees of freedom. △ Less

Submitted 29 December, 2016; v1 submitted 20 December, 2016; originally announced December 2016.

Comments: 10 pages, 8 figures. Main manuscript and supplementary information

Low temperature phase diagram of hydrogen at pressures up to 380 GPa. A possible metallic phase at 360 GPa and 200 K

Authors: M. I. Eremets, I. A. Troyan, A. P. Drozdov

Abstract: Two new phases of hydrogen have been discovered at room temperature in Ref.1: phase IV above 220 GPa and phase V above ~270 GPa. In the present work we have found a new phase VI at P~360 GPa and T<200 K. This phase is likely metallic as follows from the featureless Raman spectra, a strong drop in resistance, and absence of a photoconductive response. We studied hydrogen at low temperatures with th… ▽ More Two new phases of hydrogen have been discovered at room temperature in Ref.1: phase IV above 220 GPa and phase V above ~270 GPa. In the present work we have found a new phase VI at P~360 GPa and T<200 K. This phase is likely metallic as follows from the featureless Raman spectra, a strong drop in resistance, and absence of a photoconductive response. We studied hydrogen at low temperatures with the aid of Raman, infrared absorption, and electrical measurements at pressures up to 380 GPa, and have built a new phase diagram of hydrogen. △ Less

Submitted 18 January, 2016; originally announced January 2016.

Comments: 9 pages, 12 figures

Crystal Structure of 200 K-Superconducting Phase of Sulfur Hydride System

Authors: Mari Einaga, Masafumi Sakata, Takahiro Ishikawa, Katsuya Shimizu, Mikhail Eremets, Alexander Drozdov, Ivan Troyan, Naohisa Hirao, Yasuo Ohishi

Abstract: This article reports the experimentally clarified crystal structure of a recently discovered sulfur hydride in high temperature superconducting phase which has the highest critical temperature Tc over 200 K which has been ever reported. For understanding the mechanism of the high superconductivity, the information of its crystal structure is very essential. Herein we have carried out the simultane… ▽ More This article reports the experimentally clarified crystal structure of a recently discovered sulfur hydride in high temperature superconducting phase which has the highest critical temperature Tc over 200 K which has been ever reported. For understanding the mechanism of the high superconductivity, the information of its crystal structure is very essential. Herein we have carried out the simultaneous measurements electrical resistance and synchrotron x-ray diffraction under high pressure, and clearly revealed that the hydrogen sulfide, H2S, decomposes to H3S and its crystal structure has body-centered cubic symmetry in the superconducting phase. △ Less

Submitted 10 September, 2015; originally announced September 2015.

Comments: 8 pages, 3 figures

Superconductivity above 100 K in PH3 at high pressures

Authors: A. P. Drozdov, M. I. Eremets, I. A. Troyan

Abstract: Following the recent discovery of very high temperature conventional superconductivity in sulfur hydride (critical temperature Tc of 203 K, Ref1) we searched for superconductivity in other hydrides and found that a covalent hydride phosphine (PH3) also exhibits a high Tc >100 K at pressure P > 200 GPa as determined from four-probe electrical measurements. Following the recent discovery of very high temperature conventional superconductivity in sulfur hydride (critical temperature Tc of 203 K, Ref1) we searched for superconductivity in other hydrides and found that a covalent hydride phosphine (PH3) also exhibits a high Tc >100 K at pressure P > 200 GPa as determined from four-probe electrical measurements. △ Less

Submitted 25 August, 2015; originally announced August 2015.

Conventional superconductivity at 203 K at high pressures

Authors: A. P. Drozdov, M. I. Eremets, I. A. Troyan, V. Ksenofontov, S. I. Shylin

Abstract: A superconductor is a material that can conduct electricity with no resistance below its critical temperature (Tc). The highest Tc that has been achieved in cuprates1 is 133 K at ambient pressure2 and 164 K at high pressures3. As the nature of superconductivity in these materials has still not been explained, the prospects for a higher Tc are not clear. In contrast, the Bardeen-Cooper-Schrieffer (… ▽ More A superconductor is a material that can conduct electricity with no resistance below its critical temperature (Tc). The highest Tc that has been achieved in cuprates1 is 133 K at ambient pressure2 and 164 K at high pressures3. As the nature of superconductivity in these materials has still not been explained, the prospects for a higher Tc are not clear. In contrast, the Bardeen-Cooper-Schrieffer (BCS) theory gives a guide for achieving high Tc and does not put bounds on Tc, all that is needed is a favorable combination of high frequency phonons, strong electron-phonon coupling, and a high density of states. These conditions can be fulfilled for metallic hydrogen and covalent compounds dominated by hydrogen4,5. Numerous calculations support this idea and predict Tc of 50-235 K for many hydrides6 but only moderate Tc=17 K has been observed experimentally7. Here we studied sulfur hydride8 where a Tc~80 K was predicted9. We found that it transforms to a metal at pressure ~90 GPa. With cooling superconductivity was found deduced from a sharp drop of the resistivity to zero and a decrease of Tc with magnetic field. The pronounce isotope shift of Tc in D2S is evidence of an electron-phonon mechanism of superconductivity that is consistent with the BCS scenario. The superconductivity has been confirmed by magnetic susceptibility measurements with Tc=203K. The high Tc superconductivity most likely is due to H3S which is formed from H2S under its decomposition under pressure. Even higher Tc, room temperature superconductivity, can be expected in other hydrogen-based materials since hydrogen atoms provide the high frequency phonon modes as well as the strong electron-phonon coupling. △ Less

Submitted 26 June, 2015; originally announced June 2015.

Journal ref: Nature 525, 73 (2015)

Conventional superconductivity at 190 K at high pressures

Authors: A. P. Drozdov, M. I. Eremets, I. A. Troyan

Abstract: The highest critical temperature of superconductivity Tc has been achieved in cuprates: 133 K at ambient pressure and 164 K at high pressures. As the nature of superconductivity in these materials is still not disclosed, the prospects for a higher Tc are not clear. In contrast the Bardeen-Cooper-Schrieffer (BCS) theory gives a clear guide for achieving high Tc: it should be a favorable combination… ▽ More The highest critical temperature of superconductivity Tc has been achieved in cuprates: 133 K at ambient pressure and 164 K at high pressures. As the nature of superconductivity in these materials is still not disclosed, the prospects for a higher Tc are not clear. In contrast the Bardeen-Cooper-Schrieffer (BCS) theory gives a clear guide for achieving high Tc: it should be a favorable combination of high frequency phonons, strong coupling between electrons and phonons, and high density of states. These conditions can be fulfilled for metallic hydrogen and covalent hydrogen dominant compounds. Numerous followed calculations supported this idea and predicted Tc=100-235 K for many hydrides but only moderate Tc~17 K has been observed experimentally. Here we found that sulfur hydride transforms at P~90 GPa to metal and superconductor with Tc increasing with pressure to 150 K at ~200 GPa. This is in general agreement with recent calculations of Tc~80 K for H2S. Moreover we found superconductivity with Tc~190 K in a H2S sample pressurized to P>150 GPa at T>220 K. This superconductivity likely associates with the dissociation of H2S, and formation of SHn (n>2) hydrides. We proved occurrence of superconductivity by the drop of the resistivity at least 50 times lower than the copper resistivity, the decrease of Tc with magnetic field, and the strong isotope shift of Tc in D2S which evidences a major role of phonons in the superconductivity. H2S is a substance with a moderate content of hydrogen therefore high Tc can be expected in a wide range of hydrogen-contain materials. Hydrogen atoms seem to be essential to provide the high frequency modes in the phonon spectrum and the strong electron-phonon coupling. △ Less

Submitted 1 December, 2014; originally announced December 2014.

Scattering function for a self-avoiding polymer chain

Authors: A. D. Drozdov

Abstract: An explicit expression is derived for the scattering function of a self-avoiding polymer chain in a $d$-dimensional space. The effect of strength of segment interactions on the shape of the scattering function and the radius of gyration of the chain is studied numerically. Good agreement is demonstrated between experimental data on dilute solutions of several polymers and results of numerical si… ▽ More An explicit expression is derived for the scattering function of a self-avoiding polymer chain in a $d$-dimensional space. The effect of strength of segment interactions on the shape of the scattering function and the radius of gyration of the chain is studied numerically. Good agreement is demonstrated between experimental data on dilute solutions of several polymers and results of numerical simulation. △ Less

Submitted 6 February, 2005; originally announced February 2005.

Comments: 16 pages, 7 figures

Non-entropic theory of rubber elasticity: flexible chains with weak excluded-volume interactions

Authors: A. D. Drozdov

Abstract: Strain energy density is calculated for a network of flexible chains with weak excluded-volume interactions (whose energy is small compared with thermal energy). Constitutive equations are developed for an incompressible network of chains with segment interactions at finite deformations. These relations are applied to the study of uniaxial and equi-biaxial tension (compression), where the stress… ▽ More Strain energy density is calculated for a network of flexible chains with weak excluded-volume interactions (whose energy is small compared with thermal energy). Constitutive equations are developed for an incompressible network of chains with segment interactions at finite deformations. These relations are applied to the study of uniaxial and equi-biaxial tension (compression), where the stress--strain diagrams are analyzed numerically. It is demonstrated that intra-chain interactions (i) cause an increase in the Young's modulus of the network and (ii) induce the growth of stresses (compared to an appropriate network of Gaussian chains), which becomes substantial at relatively large elongation ratios. The effect of excluded-volume interactions on the elastic response strongly depends on the deformation mode, in particular, it is more pronounced at equi-biaxial tension than at uniaxial elongation. △ Less

Submitted 30 August, 2004; originally announced August 2004.

Comments: 21 pages, 3 figures

Non-entropic theory of rubber elasticity: flexible chains grafted on a rigid surface

Authors: A. D. Drozdov

Abstract: The elastic response is studied of a single flexible chain grafted on a rigid plane and an ensemble of non-interacting tethered chains. It is demonstrated that the entropic theory of rubber elasticity leads to conclusions that disagree with experimental data. A modification of the conventional approach is proposed, where the end-to-end distribution function (treated as the governing parameter) i… ▽ More The elastic response is studied of a single flexible chain grafted on a rigid plane and an ensemble of non-interacting tethered chains. It is demonstrated that the entropic theory of rubber elasticity leads to conclusions that disagree with experimental data. A modification of the conventional approach is proposed, where the end-to-end distribution function (treated as the governing parameter) is replaced by the average energy of a chain. It is revealed that this refinement ensures an adequate description of the mechanical behavior of flexible chains. Results of numerical simulation are compared with observations on uniaxial compression of a layer of grafted chains, and an acceptable agreement is shown between the model predictions and the experimental data. Based on the analysis of combined compression and shear, a novel micro-mechanism is proposed for the reduction of friction of polymer melts at rigid walls. △ Less

Submitted 22 August, 2004; originally announced August 2004.

Comments: 16 pages, 2 figures

Stiffness of polymer chains

Authors: A. D. Drozdov

Abstract: A formula is derived for stiffness of a polymer chain in terms of the distribution function of end-to-end vectors. This relationship is applied to calculate the stiffness of Gaussian chains (neutral and carrying electric charges at the ends), chains modeled as self-avoiding random walks, as well as semi-flexible (worm-like and Dirac) chains. The effects of persistence length and Bjerrum's length… ▽ More A formula is derived for stiffness of a polymer chain in terms of the distribution function of end-to-end vectors. This relationship is applied to calculate the stiffness of Gaussian chains (neutral and carrying electric charges at the ends), chains modeled as self-avoiding random walks, as well as semi-flexible (worm-like and Dirac) chains. The effects of persistence length and Bjerrum's length on the chain stiffness are analyzed numerically. An explicit expression is developed for the radial distribution function of a chain with the maximum stiffness. △ Less

Submitted 28 July, 2004; originally announced July 2004.

Comments: 21 pages, 6 figures

The end-to-end distribution function for a flexible chain with weak excluded-volume interactions

Authors: A. D. Drozdov

Abstract: An explicit expression is derived for the distribution function of end-to-end vectors and for the mean square end-to-end distance of a flexible chain with excluded-volume interactions. The Hamiltonian for a flexible chain with weak intra-chain interactions is determined by two small parameters: the ratio $ε$ of the energy of interaction between segments (within a sphere whose radius coincides wi… ▽ More An explicit expression is derived for the distribution function of end-to-end vectors and for the mean square end-to-end distance of a flexible chain with excluded-volume interactions. The Hamiltonian for a flexible chain with weak intra-chain interactions is determined by two small parameters: the ratio $ε$ of the energy of interaction between segments (within a sphere whose radius coincides with the cut-off length for the potential) to the thermal energy, and the ratio $δ$ of the cut-off length to the radius of gyration for a Gaussian chain. Unlike conventional approaches grounded on the mean-field evaluation of the end-to-end distance, the Green function is found explicitly (in the first approximation with respect to $ε$). It is demonstrated that (i) the distribution function depends on $ε$ in a regular way, while its dependence on $δ$ is singular, and (ii) the leading term in the expression for the mean square end-to-end distance linearly grows with $ε$ and remains independent of $δ$. △ Less

Submitted 25 June, 2004; originally announced June 2004.

Comments: 39 pages, 1 figure

Networks of self-avoiding chains and Ogden-type constitutive equations for elastomers

Authors: A. D. Drozdov, M. Gottlieb

Abstract: An expression is derived for the strain energy of a polymer chain under an arbitrary three-dimensional deformation with finite strains. For a Gaussian chain, this expression is reduced to the conventional Moony--Rivlin constitutive law, while for non-Gaussian chains it implies novel constitutive relations. Based on the three-chain approximation, explicit formulas are developed for the strain ene… ▽ More An expression is derived for the strain energy of a polymer chain under an arbitrary three-dimensional deformation with finite strains. For a Gaussian chain, this expression is reduced to the conventional Moony--Rivlin constitutive law, while for non-Gaussian chains it implies novel constitutive relations. Based on the three-chain approximation, explicit formulas are developed for the strain energy of a chain modeled as a self-avoiding random walk. In the case of self-avoiding chains with stretched-exponential distribution function of end-to-end vectors, the strain energy density of a network is described by the Ogden law with only two material constants. For the des Cloizeaux distribution function, the constitutive equation involves three adjustable parameters. The governing equations are verified by fitting observations on uniaxial tension, uniaxial compression and biaxial tension of elastomers. Good agreement is demonstrated between the experimental data and the results of numerical analysis. An analytical formula is derived for the ratio of the Young's modulus of a self-avoiding chain to that of a Gaussian chain. It is found that the elastic modulus per chain in the Ogden network exceeds that in a Gaussian network by a factor of three, whereas the elastic modulus of a chain with the generalized stretched exponential distribution function equals about half of the modulus of a Gaussian chain. △ Less

Submitted 12 May, 2004; originally announced May 2004.

Comments: 37 pages, 13 figures

Constitutive equations for a polymer fluid based on the concept of non-affine networks

Authors: A. D. Drozdov, R. K. Gupta

Abstract: Constitutive equations are developed for a polymer fluid, which is treated as a permanent network of strands bridged by junctions. The junctions are assumed to slide with respect to their reference positions under loading. Governing equations are derived by using the laws of thermodynamics under the assumption that the vorticity tensor for the flow of junctions is proportional to that for macro-… ▽ More Constitutive equations are developed for a polymer fluid, which is treated as a permanent network of strands bridged by junctions. The junctions are assumed to slide with respect to their reference positions under loading. Governing equations are derived by using the laws of thermodynamics under the assumption that the vorticity tensor for the flow of junctions is proportional to that for macro-deformation. Explicit expressions are developed for the steady elongational viscosity, as well as for the steady shear viscosity and normal stress functions. To verify the constitutive relations, three sets of experimental data are approximated on polystyrene solutions with various molecular weights. It is demonstrated that the model can correctly describe stress overshoot for the shear stress and first normal stress difference in start-up tests with various strain rates. Adjustable parameters in the governing equations change consistently with the strain rate, molecular weight and concentration of entanglements. To validate the constitutive equations, observations on low-density polyethylene melt in uniaxial extensional flow are compared with the results of numerical analysis when the material constants are found by matching experimental data in shear tests. △ Less

Submitted 18 February, 2004; originally announced February 2004.

Comments: 40 pages, 12 figures

Linear viscoelasticity of polyolefin melts: the effects of temperature and chain branching

Authors: A. D. Drozdov, S. Agarwal, R. K. Gupta

Abstract: Observations are reported in isothermal torsional oscillation tests on melts of isotactic polypropylene (iPP) and low-density polyethylene (LDPE) in the intervals of temperature between 190 and 250C (iPP) and between 120 and 190C (LDPE). With reference to the concept of transient networks, constitutive equations are developed for the viscoelastic response of polymer melts at three-dimensional de… ▽ More Observations are reported in isothermal torsional oscillation tests on melts of isotactic polypropylene (iPP) and low-density polyethylene (LDPE) in the intervals of temperature between 190 and 250C (iPP) and between 120 and 190C (LDPE). With reference to the concept of transient networks, constitutive equations are developed for the viscoelastic response of polymer melts at three-dimensional deformations with small strains. A melt is treated as an equivalent network of strands bridged by temporary junctions (entanglements and physical cross-links whose life-times exceed the characteristic time of deformation). The time-dependent behavior of the network is modelled as detachment of active strands from their junctions and merging of dangling strands with the network. The network is assumed to be strongly heterogeneous in the sense that different junctions have different activation energies for separation of strands. The stress-strain relations involve three adjustable parameters (the plateau modulus, the average activation energy for rearrangement of strands and the standard deviation of activation energies) that are determined by matching the dependencies of storage and loss moduli on frequency of oscillations. The difference in the effects of temperature on the material constants of iPP and LDPE is associated with the difference in their molecular architecture. △ Less

Submitted 13 November, 2003; originally announced November 2003.

Comments: 20 pages, 8 figures

Thermal degradation and viscoelasticity of polypropylene-clay nanocomposites

Authors: A. D. Drozdov, A. Al-Mulla, D. A. Drozdov, R. K. Gupta

Abstract: Results of torsional oscillation tests are reported that were performed at the temperature T=230C on melts of a hybrid nanocomposite consisting of isotactic polypropylene reinforced with 5 wt.% of montmorillonite clay. Prior to mechanical testing, specimens were annealed at temperatures ranging from 250 to 310C for various amounts of time (from 15 to 420 min). Thermal treatment induced degradati… ▽ More Results of torsional oscillation tests are reported that were performed at the temperature T=230C on melts of a hybrid nanocomposite consisting of isotactic polypropylene reinforced with 5 wt.% of montmorillonite clay. Prior to mechanical testing, specimens were annealed at temperatures ranging from 250 to 310C for various amounts of time (from 15 to 420 min). Thermal treatment induced degradation of the matrix and a pronounced decrease in its molecular weight. An integro-differential equation is derived for the evolution of molecular weight based on the fragmentation-aggregation concept. This relation involves two adjustable parameters that are found by fitting observations. With reference to the theory of transient networks, constitutive equations are developed for the viscoelastic response of nanocomposite melts. The stress-strain relations are characterized by three material constants (the shear modulus, the average energy for rearrangement of strands and the standard deviation of activation energies) that are determined by matching the dependencies of storage and loss moduli on frequency of oscillations. Good agreement is demonstrated between the experimental data and the results of numerical simulation. It is revealed that the average energy for separation of strands from temporary junctions is independent of molecular weight, whereas the elastic modulus and the standard deviation of activation energies linearly increase with mass-average molecular weight. △ Less

Submitted 3 October, 2003; originally announced October 2003.

Comments: 24 pages and 18 figures

Kinetic equations for thermal degradation of polymers

Authors: A. D. Drozdov

Abstract: Kinetic equations are analyzed for thermal degradation of polymers. The governing relations are based on the fragmentation-annihilation concept. Explicit solutions to these equations are derived in two particular cases of interest. For arbitrary values of adjustable parameters, the evolution of the number-average and mass-average molecular weights of polymers is analyzed numerically. Good agreem… ▽ More Kinetic equations are analyzed for thermal degradation of polymers. The governing relations are based on the fragmentation-annihilation concept. Explicit solutions to these equations are derived in two particular cases of interest. For arbitrary values of adjustable parameters, the evolution of the number-average and mass-average molecular weights of polymers is analyzed numerically. Good agreement is demonstrated between the results of numerical simulation and experimental data. It is revealed that the model can correctly predict observations in thermo-gravimetric tests when its parameters are determined by matching experimental data for the decrease in molecular weight with exposure time. △ Less

Submitted 29 September, 2003; originally announced September 2003.

The effect of strain rate on the viscoplastic behavior of isotactic polypropylene at finite strains

Authors: A. D. Drozdov, J. deC. Christiansen

Abstract: Two series of uniaxial tensile tests are performed on isotactic polypropylene with the strain rates ranging from 5 to 200 mm/min. In the first series, injection-molded specimens are used without thermal pre-treatment, whereas in the other series, the samples are annealed for 51 h at 160C prior to testing. A constitutive model is developed for the viscoplastic behavior of isotactic polypropylen… ▽ More Two series of uniaxial tensile tests are performed on isotactic polypropylene with the strain rates ranging from 5 to 200 mm/min. In the first series, injection-molded specimens are used without thermal pre-treatment, whereas in the other series, the samples are annealed for 51 h at 160C prior to testing. A constitutive model is developed for the viscoplastic behavior of isotactic polypropylene at finite strains. A semicrystalline polymer is treated as an equivalent heterogeneous network of chains bridged by permanent junctions (physical cross-links and entanglements). The network is thought of as an ensemble of meso-regions connected with each other by links (lamellar blocks). In the sub-yield region of deformations, junctions between chains in meso-domains slide with respect to their reference positions (which reflects sliding of nodes in the amorphous phase and fine slip of lamellar blocks). Above the yield point, sliding of nodes is accompanied by displacements of meso-domains in the ensemble with respect to each other (which reflects coarse slip and fragmentation of lamellar blocks). Stress-strain relations for a semicrystalline polymer are derived by using the laws of thermodynamics. The constitutive equations are determined by 5 adjustable parameters that are found by matching observations. Fair agreement is demonstrated between the experimental data and the results of numerical simulation. △ Less

Submitted 31 July, 2002; originally announced July 2002.

Comments: 27 pages, 20 figures