Our tools for computational materials simulation

For the purpose of performing computational materials simulations (CMS), one very essential factor is the availability of some highly specialised computational tools. Not only we need these tools, equally important is the know-how to use these tools. When I mention 'tools' I have in my mind hardware and software tools.

Hardware resources we have in the physics school are 'moderately abundant'. We have cluster PCs and workstations, totaling up to around 256 cores. The main purpose of these hardware is to provide processor cores for intensive computational purposes. For better performance, these computers have to also equipped with good amount of RAM. In the long term, when the group's size expands, the available hardware resources may become very tight. But at the moment we are quite comfortable with what we have.

Software tools can further be broken down into two levels of complexity: (1) running the software, and (2) performing post-processing on the data churned out by these software.

The software tools that are available to us include: ABINIT, Wien2k, DeMon, LAMMPS, DFTB+, HOTBIT. We also have licensed software VASP and Crystal, but we have not been very skillful in using these compared to the former. We also have experienced with Gaussian and Material Studio, but unfortunately we do not has full access to the licensed version of these. All these software mentioned above (except VASP and Crystal) have been fully installed in our hardware. Installation and running VASP and Crystal are not a priority for the moment as we have more software than we can digest.

There are basically three major categories of software code for the purpose of CMS. DFT software (ABINIT, Gaussian, VASP, Crystal, DeMon), tight-binding DFT (DFTB+, HOTBIT), and molecular dynamics software (LAMMPS). Basically we have installed these software in our computational resources (most are rendered to run in parallelised modes). And most importantly, we have (I mean our group members) gained enough experience to use them (except for VASP and Crystal).

However, when running CMS calculation, very often we need to know what are the ground state (GS) structures at zero temperature, i.e., configurations that correspond to the lowest energy. If a system is to be stable (hence corresponds to a state observable in the lab), it should be a GS, or a meta stable state (which is not as stable, and is subjected to transition to other nearby meta-stable states upon thermal perturbation). Finding GS structure by itself is a difficult task, since there are in principle almost infinitely many possible ways how a system can be configured in a finite 3D space. To this end, we need some intelligent ways to help us finding GS. This can be achieved via genetic algorithm (GA) and basin hoping (BH). Most often, one has to couple GA and/or BH into the above software to obtain the GS of a system. BH and GA programs for locating GS purposes are usually standalone codes developed by individual research group. However, GASP, a GA program has recently been developed by the Cornell materials engineering group, and implemented into LAMMPS. Although the GASP+LAMMPS have been configured to run in our computer nodes, we have yet to develop the know-how to use them for our GS-finding purpose. This is due to the reason that we have not had the time to explore the GASP codes and to tune the GA parameters.

A GA (or BH) coupled to LAMMPS, or DFTB+, allows one to search for the GS at zero temperature. Without GA or BH (or other global-minimisation search algorithms) one can not confidently claim a given configuration of atoms is in a stable state. A calculation for the total energy of such an arbitrary configuration at zero temperature cannot be convincingly claimed to be the global minimum. Having a workable GS-searching capability (which has to be coupled to the CMS software) is what we have been trying to acquire for sometimes. The good news is we have just acquired such an "upgrade" recently. Well, we did not do it ourselves. All credit goes to Prof. Lai's group in the NCU complex fluid research lab. Prof. Lai's group has developed a highly effective parallelised GA code that could be easily coupled to LAMMPS and DFTB+. On top of this, they also has a BH code that could also be used in place of GA as a stand-alone optimisation tool. We simply borrow their achievement which they developed with tears and blood. The next thing we want to develop is to couple the GA and BH with DFT software. Technically it is possible to implement such a software-coupling. But we also anticipate a  serious penalty on the speed which is hampered by the expensive calculation cost of the DFT codes.

In most cases, all output from the CMS calculations (be it DFT, DFTB+ or LAMMPS) have to be post-processed. And usually the post-processing of these data have to be manually done. The software packages usually provide very limited functionality for post-processing, except some very common ones. For example, plotting the electronic density from DFT calculation is a norm, hence can be easily done by using the post-processing packages come together with, say, Gaussians' gaussiview. However, for MD and DFTB+, almost no such post-processing tools exist. We have to write out own codes to perform post-processing. For example, we have to write our own Mathematica code to calculate the average bond length of certain selected atomic species in the LAMMPS output data. To calculate the point group of a bunch of Boron atoms in a cluster in a typical DFTB+ output, we have to manually do this using an independent software that calculate point group based on the coordinates of the atoms.

Of course, there are a lot of  potential technical issues that could hinder us from completing the intended calculation. The most serious hindrance comes in the form of the unavailability of force fields, or unsuitability of the force fields (in the case of LAMMPS), or SK files (in the case of DFTB). In DFT calculation, there is no issue of forcefields or SK files, but the problems in DFT are non-convergence, or simply taking too long hours to complete the calculation.

After the calculations have completed, then we have another kind of issue to address, namely, the reliability of the results, and the physical interpretation of these results. The former are usually easier than the later. We need to know a lot of physics in order to write a meaningful sentence to address the physical interpretation of the results. This is the part we have to keep improving and learning from Prof. Lai. In comparison, running calculation is relatively easier then assuring the correctness of the calculation, or interpreting the physics of the results. 

The point I wish to highlight is that, after an extended period of hard work, we have gained a lot of experience in installing, maintaining, using and manipulating these specialised software mentioned above. These technical know-how are by no means easy, but have to be earned via many hours of learning and effort. Once we have identified a problem to attack, we are now in a better position to go ahead confidently. In order for a new comer to perform a CMS, he/she will have to master all these software tools mentioned above, apart from knowing the physics and the theories. Even if she/he can't be a great CMS scientist, at least she/he will be a well-trained computer expert in Linux, post-processing, programming, and visualisation.

Some thought on my CMS research

I must admit that at the moment the objective of my computational materials science (CMS) plan is still diffusive, especially the part on 'synergistic collaboration' with the experimental colleagues. To sort out a working model to conduct such kind of collaboration is still at its preliminary stage. At the moment only two such collaborations have bear results, which are the ZnO simulation paper with Dr. Sharom's group and the one with Nazarov (on DFT calculation of phosphors).

There is no lack of research topics for us at the moment, but I suppose our calculation projects are still a bit shallow and not making enough impact (with an exceptional case - the forthcoming epitaxial graphene paper could be very significant, thanks to Prof. Lai's insight). I recall that in TU Delft (in Holland), people are working on mesoscopic physics calculations in collaboration with the experimental team downstairs. The projects are intensive, well defined and very specific, and often lead to publication in Science or Nature. Maybe for my case i can't really demand that much as our experimentalists can do only that much and that deep. Typically they are good at characterising a piece of materials through a routine protocol, or cooking a sample from the oven through a standard  procedure. It is quite difficult for them to 'control' or tune their experiment's parameter (e.g. changing the concentration of mixture in an alloy or the atomistic thickness in their samples, or even just changing the viewing angle when measuring the optical spectrum of a sample using the machines, etc). Usually they have only one or two samples bought using grant money. If fortunate they will have five samples, of which two are accidentally broken. Samples cooked / synthesised by them are usually coarse, and the quality is not uniform or guaranteed. Often they can't control the samples' properties accurately. All these make the modelling of these systems a difficulty. It is often difficult to identify the "empty slot" where atomistic computational calculations can come in.

Well, not all experiments by our experimentalists can't be modelled. One such successful case is the
MD modellig of the ZnO paper. It is to some extent an 'accident' discovered by JJ. When it comes to the paper writing part I have to think very hard to 'merge' their experimental findings with our MD results.

There is currently one project to calculate ternary alloy's vibrational mode and compare it with experiments. This is a PhD project to be conducted by Pauline. However it is still too early to say anything about the success or failure of it as the student has not really began the DFT calculation yet.

In another case, an African student came to talk to me once. i like his initiative and proactiveness. He has an experimental result in which he synthesised iron hydroxide nanoparticle and used it to prevent blood sample from clogging. This experiment, although still coarse, is possible to be modeled using MD. But it may involve chemistry and chemical reaction. This is a daunting system for me, so i do not dare to promise anything. I told the Nigerian student that I will come back to him only when I am more ready. Hope that this day will finally come true.

So for the moment there is no lack of project to run, and we will be kept occupied for quite an extended period of time. Just that the problems we are working on is mostly 'shallow' and static (e.g. ground state structures, thermal properties, DOS and things like that). Of course I am not feeling bad about this because I totally understand that as new comers to the field we must begin from a simple case before hoping to a more advanced 'dynamical' systems, i.e. that involve chemical reaction, reconstruction or transport. In this sense, Jing Qiang's presence in this world becomes important to me. I suppose he will be working on what I described as 'dynamic' systems, and he will share with me the details and the concerns one has to take into consideration when modelling them, the techniques, and theories and the tools. If Jing Qiang can tell me in all the details of how to conduct the modelling of such complicated system, i believe that surely will kick start our local research to the next level of height in CMS. (JQ is a currently a P.hD. student working on CMS in Finland now, also a very closed and dear friend of mine).

Priority misplacement in our local university

USM is entering its second phase as an APEX university. We was told from the first APEX phase to  submit any proposal that could elevate the university status in terms of research output. However bulky bureaucratic procedures and the lack of information channel have not contributed in making the proposal submission by any poorly-contacted academic staff a task worth trying. At the back of this reluctance is the prototypical opinion that any suggestion will either fall into deaf ears or results in lip service responses. I think I have tried to make a few informal attempts there and then some proposals to improve the academic quality in the university, but I do not register any memory that these attempts ever bear results that make a marked difference. Overall, there has been little change in the way how business is done despite the phrase 'business unusual' was the slogan of the day since the university acquired the APEX status many years ago. 

As a researcher in the school's academic team I am free to carry out my own research very freely without interruption from the university. This is taken by me as a deep blessing. The university does provide much freedom to their academics to conduct their own research. But on the other hand the university has never been actively improving the academic environment to makes the university a more conducive place for quality research. The primary root cause of the lack of conducive academic environment is structural: that bureaucrats dominate over academicians (not only in our university but in fact in all our local universities). Whenever anyone has a proposal to implement something new (e.g. a new course, a new way to handling things, a new facility, a new system ...), the first consideration by the people in charge is not to think of how to actualise such proposal. Instead their first thought will be: is this allowed by the existing rules? Such an inappropriate priority has a deep negative impact in creating a top class university and an innovative ambiance. 

There is nothing wrong about obeying rules. The main issues is people are deeply mired in the lethargic thinking that we must not change the rule but to follow them. People never ask themselves this question: are not the rules created to make things work better? If these rules become counterproductive, should not they be reconsidered, modified or just discarded? No. People don't ask such question. They just follow the status quo, which is the most comforting thing to do, and also the safest. 

Most often rules in the university are used as the most convenient pretext to turn down a innovative proposal. For example, can we hire a half-paid researcher and allow him/her to work as an paid researcher in another university? Not so, I suppose, even if he is the first Malaysia having the potential to win a Nobel prize. How about setting up a new course that is deemed most popular in the market? No, you can't simply do that. It takes at least half a year to go through all the necessary procedures before a new course is approved and launched. Can't we just lunch a new course in two weeks time as long as an expert has made all necessary preparation to teach the course? In principle yes, but never in reality. In some overseas universities, as long as an expert in the field wills, he can at anytime offer a new course teaching the most up-to-date findings in his field. So in this specific case, demanding a new course to fulfill all the bureaucratic procedures of getting approval is actually counterproductive. But yet people keep on insisting that fulfilling the rules is more important that delivering new knowledge. If the people in charge is proactive enough it is possible to   place academic priority over bureaucratic consideration. I have not seen such thing happen yet. 

Giving priority to the rules rather than real productivity has blocked the way for excellence in our local university. Somewhat pessimistically I do not expect in our life time that such bureaucratic-dominated mentality will see any sea change.

My proposal for a computational materials science expertise unit in the physics school

Below is my response to the call for proposal for the Pelan Pelaksanaan Apex Fasa 2 (Work Plan 2014-2016) by the physics school. The proposal is basically a plan of my own ambition to explore the field of computational condensed matter physics in USM. I do not really expect any positive outcome from this. Based on previous experience the fate of the proposal will most probably be just like the others we submitted in the last few years. Despite my pessimistic anticipation, I have spent my precious time and effort to prepare a serious proposal. It implicitly been encoded with my research direction and aspiration. Research has always been one of the two prime priorities of my life. The proposal reflects from a certain view point my personal plan to excellence. As a matter of fact I don't  really care whether the proposal will be shoot down during decision making by the school or just be partially approved. It is just a symbolic move in which I am reinforcing a promise to myself, and an ambition to be achieved in the long term.



Title:Establishment of computational materials science expertise in the School of Physics, USM

Objectives:

1. To establish a strong computational materials science research team specialised in solving condensed matter physics and materials science problems using high performance computers (HPC).

2. To establish a synergistic research collaboration between the Theoretical and Computational Physics group with other experimentalists within USM to theoretically investigate, interpret and design of novel material systems via computational simulations.

Motivation:
Computational condensed matter physics (used interchangeably with computational materials science in our context) is an interdisciplinary research area which main objectives include understanding, modeling, predicting, and ultimately, theoretically engineering the physical properties of realistic materials. In this research discipline, state-of-the-arts computational techniques are the weapons employed to simulate the physics of material systems at atomistic level based on quantum-mechanical or semi-empirical prescriptions. To this end, known therories in condensed matter physics, chemistry and computer science are combined and applied numerically. Prediction of macroscopic properies of materials via such approach is made possible by the spectacular increase in computational power and novel numerical algorithms, allowing fundamental equations governing the physics at the atomic level to be solved numerically and with unprecedented accuracy. Today, based only the knowledge of a single atom, one can predict how the material formed by that atom type will look, what properties that material will have and how it will behave under certain conditions. By simply changing the arrangement of constituent atoms, or by adding atoms of a different type, the macroscopic properties of all materials can be modified. It is in this way that one can learn how to improve mechanical, optical and/or electronic properties of known materials, or one can predict properties of new materials, those which are not found in nature but are designed and synthesized in the laboratory. High performance computers (HPC) and routine visualization software are utilized to generate direct comparisons with experimental conditions.

Computational materials science has also the great advantange to compliment and guide experimental searches. This point is particularly relevant to our local solid-state laboratories, where novel materials are routinely synthesised and characterised through all sorts of experimental tools, but rarely complemented by computational simulations at atomistic level. Apart from its capability to provide theoretical insight at the microscopic level for the physical origin of a condensed matter system, computational material techniques also provide a very wide oppertunity to publish relatively easily in international journals, at a relatively low cost (because it uses only computing power of computers as its main ‘ingredient’ rather than relying on real materials and experiment hard ware facilities).

Synergistic collaboration between experimentalists and theorists is not a common practice in USM, especially in the School of Physics. There are many experimental results from the NOR lab and X-ray lab that can be complemented by molecular dynamics or other computational techniques. For example, the structural phase transition observed in phenol-amines adducts is in principle model-able using either monte carlo or molecular dynamics. The XRD spectrum or Raman spectrum on GaN or ZnO samples can be simulated by constructing supercell models with density functional theory (DFT) codes. So are those novel X-ray cyrstallographic structures of organic crystal solved routinely in the X-ray lab can be calculated using DFT or MD codes. Motivated by the existence of such a huge opportunity of existing in-house experimental resources, we propose an effort to tap the potentiality and translate it into real publication and quality research outcome: by establishing a strong computational physics expertise in the School of Physics, USM.

In practice computational materials science research requires most importantly the technical know-how to carry out the computational tasks and demands relatively cheap monetary cost (mostly for setting up computer hardware, and to a lesser degree, purchase of software). However, as far as we are aware of, this is a research field barely practiced in Malaysia despite its obvious practical advantages. This is presumably due to the lack of specialised training and experts in this area, apart from its high threshold (in terms of technical knowledge) to enter the field. We are now among the very rare species in Malaysia that are able to perform atomistic computational materials simulation using highly specialised software and HPC, e.g., cluster Linux computing system, highly parallelised codes, e.g., LAMMPS, Wien2k, Gaussian, DFTB+, quantum Monte Carlo and genetic algorithm codes. Thanks to the research experiences accumulated throughout the years on computational condensed matter systems, we (the members in theoretical and computational physics group) have now readily equipped with the technical know-how to apply these specialised computational skill to perform calculations on real materials.

Having established the potential, practical advantages of and our readiness in this research front, we propose to the School of Physics to strategically create an expert team for computational condensed matter physics. The team can be formally considered as a subgroup under the theoretical physics and computational physics research group. The subgroup will team up with experimentalists from all the research labs in USM, especially those from the School of Physics, to form a synergic research collaboration in which experimental investigations of novel materials are coupled with state-of-the-arts computational physics techniques. For the sake of reference, we shall refer to this subgroup as the computational materials science expertise unit, or just the expertise unit, hereafter. The proposed expertise unit will develop all the necessary expertise, in particular first-principles calculation and molecular dynamics methods, to compliment the research investigations carried out in our experimentalist colleagues' lab.

Current status:
Leader
Computational condensed matter physics research in the School of Physics, USM, is first pioneered by Dr. Yoon Tiem Leong, in closed collaboration with experts in the field from MMU Melaka, National Taiwan National University and Academy Sciences of Moldova.

Publications (already published)
1. Thong Leng Lim, Mihail Nazarov, Tiem Leong Yoon, Lay Chen Low, M. N. Ahmad Fauzi, X-ray diffraction experiments, luminescence measurements and first-principles GGA+U calculations on YTaO4, Computational Materials Science 77 (2013) 13–18 (http://dx.doi.org/10.1016/j.commatsci.2013.03.042).

2. Wen Fong Goh, Sohail Aziz Khan and Tiem Leong Yoon, A molecular dynamics study of the thermodynamic properties of barium zirconate, Modelling Simul. Mater. Sci. Eng. 21 (2013) 045001 (11pp).

3. Molecular dynamics simulation of thermodynamic and thermal transport properties of strontium titanate with improved potential parameters, GOH Wen Fong, YOON Tiem Leong, Sohail Aziz KHAN, Computational Material Science 60 (2012) 123–129.

4. Surface and interface phonon polaritons of wurtzite GaN thin film grown on 6H-SiC substrate, S. S. Ng, T. L. Yoon, Z. Hassan, and H. Abu Hassan, Applied Physics Letters 94, 241912 (2009).

5. Yoon Tiem Leong, Goh Eong Sheng, Calculation of ground state energy of a “4 × 4” flux qubit Josephson junction array using diffusion quantum Monte Carlo Method, PERFIK 2012, Bukit Tinggi, Pahang, Malaysia, 21 Nov 2012 (AIP Conf. Proc. 1528, pp. 384-389; doi: http://dx.doi.org/10.1063/1.4803631).

Papers submitted for publication in peer reviewed journals:
6. Thong Leng Lim, Mihail Nazarov, Tiem Leong Yoon, Lay Chen Low, M. N. Ahmad Fauzi, Ab initio calculations and luminescence study of YNb$O_4$ (Scripta Physica, submitted)

7. Tjun Kit Min, Tiem Leong Yoon , Chuo Ann Ling, Shahrom Mahmud, Thong Leng Lim, Annealing of ZnO surfaces via molecular dynamics simulation with reactive force field (Surface Science, submitted)

Papers in preparation for publication in peer reviewed journals
8. Epitaxial growth of graphene on 6H-silicon carbide substrate by simulated annealing method (in collaboration with S. K. Lai, NCU Taiwan, in preparation).

9. Temperature Quench Molecular Dynamics Simulation of Phase Coexistence Curve of Lennard-Jones Fluid (Goh Eong Sheng, Yoon Tiem Leong, in preparation).

Current research projects:
1. Molecular dynamics simulation of epitaxial graphene growth

2. Genetic Algorithm assisted DFTB calculations on boron clusters

3. DFT calculations on new generation of phosphors

4. Molecular dynamics simulation of graphene nanoribbon melting

5. Molecular dynamics simulation of carbon nanotube melting

6. DFT calculation on ferroelectrics (Ph.D project)

7. DFT calculation of phonon vibrational modes in ternary alloy (Ph.D project, collaboration with experimentalist from NOR lab).

8. 3D FDTD Modeling of the effects of electromagnetic phenomena in the ionosphere and Earth’s magnetic field over the Sumatera-Malaysia region (Ph.D project, in collaboration with remote sensing group)

Research students
1. Ng Wei Chun, research assistant (RA) – already obtained M.Sc offer latter from USM, to register soon.

2. Min Tjun Kit, research assistant (RA) – already obtained M.Sc offer latter from USM, to register soon.

3. Siti Harwani bt Md Yusoff, current Ph.D student.

4. Lee Thong Yan, current Ph.D student.

5. Pauline Yeoh, current Ph.D student.

6. Goh Wen Fong (M.Sc, graduated).

Research Collaborators
1. Dr.  Lim Thong Leng, Faculty of Engineering and Technology, Multimedia University (Melaka), Malaysia.

2. Prof. S. K. Lai, National Central University, Taiwan.

3. Prof. Mihail Nazarov, Institute of Applied Physics, Academy Sciences of Moldova, Republic of Moldova.

4. Dr. Shahrom Mahmud, NOR lab, USM (experimentalist)

5. Dr. Ng Sha Shiong, NOR lab, USM (experimentalist)

6. Dr. Saw Kim Guan, PPJJ, USM (experimentalist)

Current computing resources
(i) Hardware
We have more than a combined number of 256 cpu cores available for HPC parallel computing. All of these hardware were build from ground zero with our own effort, and are currently maintained also by ourselves (cooling systems, LAN connections, software and hardware technical problems, etc.) with technical consultation provided by (1) Mr. Tan Choo Jun, a doctorate student from School of Computer Science, USM, and (2) Associate Prof. Dr. Chan Huah Yong of the School of Computer Science, USM. Hardware resources available to group members and members from the theoretical physics group are listed below:

1. comsics cluster (comsics.usm.my, 20 nodes x 4 intel i5 cores, Linux Rocks OS). Located in the Integrated Computater Lab, 3rd floor, Physics School building.

2. anicca cluster (anicca.usm.my, 20 nodes x 4 intel core 2 duo, Linux Rocks OS). Located in the Integrated Computater Lab, 3rd floor, Physics School building.

3. jaws workstation (Supermicro workstation, 64 x AMD 2.2 GHz Interlagos cores, CENTOS 6.3 OS). Located in the server room "Bilik Delta" in student center, 2nd floor, Physics School building.

4. chakra cluster (4 nodes x 8 intel i7 cores, 3.4 GHz, CENTOS 6.3 OS). Located in the Theory Lab, 3rd floor, Physics School building.

All these HPC hardware are capable of running MPI-enabled parallel computing. The comsics and anicca cluster are built by using PCs in the integrated computer lab, School of Physics. Formally the computers in the computer lab belongs to the School of Physics for the purpose of running Computational Physics ZCE 111 and MAT 181 courses, and occasionally, conducting workshop. The computer clusters in the computer lab are used to run various computational simulations during free period (i.e., when no classes / workshop are being conducted in the computer lab). In this sense the computers in the computer lab are being optimally ultilised for both teaching and research purposes without noticeable interference between these two modes of usage.

(ii) Software
Software installed in our HPC resources include:
Mathematica (fully licensed), Wien2k (DFT package, fully licensed), CRYSTALS (DFT package, fully licensed), VASP (DFT package, fully licensed), LAMMPS (MD package, free), DFTB+ (DFTB package, free), genetic algorithm codes, basin hoping codes (both codes are meant for finding global minimisation purposes, developed by NCU group from Taiwan).

Implementation plan
Who will be involved
1. The key player will be the existing expert in computational physics in the theoretical and computational physics group, Dr. Yoon. He will be responsible for the operation of the expertise unit, along with his graduate students, project students and research assistants.
2. Other theoretical and computational physics group members who are interested in running computational simulations or HPC calculations.
3. All researchers and experimentalists in the School of Physics or from other schools within USM (e.g. PPJJ, Materials Engineering, Chemistry School, etc.) who are interested to incorporate an intensive computational component in their researches are all welcomed. Graduate students from other research labs are encouraged to engage in collaborative project with the proposed expertise unit.

Main activities
1. Conducting high-impact researches in computational physics / computational condensed matter physics / computational materials science.

2. Creating international and national research linkages in the field, including inviting experts from the related field to the school of physics for research visit from time to time.

3. Training of our own human capital in the research field of computational condensed matter physics.

4. Providing training to current researches / students / research personnel on advanced level computing techniques and computational methods for generic purposes (such as programming in Mathematica, Fortran, GPU programming, parallel programming, visualization techniques, software and computer system maintenance, Linux, LaTeX, virtualization, etc.)

5. Acting as technical consultation and service provider entity for the physics school research community as a whole on issues related to HPC and other computationally related issues, such as setting up of parallel computing facilities, purchasing of high-performing computing facilities, installing software in Linux systems, etc.

6. Training of graduate students / academic staffs / research personnel from other research labs to run numerical programming prompted by their specific research needs.

Human resource
In order to effectively realize the ideal role of the expertise unit as proposed above, the single most important factor is to train our own experts. At the moment the only expert in the field is Dr. Yoon. But the proper functioning of the expert unit definitely necessitates more human power. To this end the School of Physics should

1. provide practical incentive and encouragement for undergraduate and graduate students to take up projects / courses provided by the expertise unit. This incentive could be in the form a guaranteed scholarship for Ph. D or M.Sc. students taking up projects in computational materials science research.

2. provide monetary incentive to graduate students from the experimental research lab to incorporate computational component in their researches.

3. tenure a new academic staff in the field of computational condensed matter physics or computational physics.

4. set up a post-doc position for computational condensed matter physics or computational physics.

5. encourage existing academic staff to incorporate more computational physics component in their research.

6. provide financial allocation to invite internationally renowned experts in the research field to visit School of Physics for an extended period or time, or even to conduct short courses.

7. The expertise unit does not request for any non-academic staff.


Computing hardware and software requirement
Rack mount computer cluster
The only hardware needed for computational condensed matter physics is computers, the more the better. As a teaching lab, the current computing facilities in the computer lab are not especially built for running really huge computational job. For a start we recommend to purchase a scalable rack-mount computer cluster system. This kind of computing facility is flexible, easy to maintain, relatively cheap and compact (less space consuming). The system comprise of a metal rack (of the size of a fridge) into which one could slot in a number of ‘blades’, where each blade is a plug-and-play motherboard having several slots of multiple-core processor. Depending on the availability of funding, the computational power of the system can be upgraded from time to time. This shall be the “primary weapon” of the expertise unit to tackle computational research problems.

Software
Software wise, many major software for computational physics are either open source (e.g. ABINIT, LAMMPS, DFTB+, CPMD, etc.) or free for academic use (e.g. Intel Fortran and its libraries). We do not expect to spend much on software as is for computer hardware.

Maintenance
1. The proposed expertise unit would be fully responsible to maintain all the computer hardware, software, and the physical spaces in which the computers are sitting. A small budget should be allocated annually for maintenance purposes, such as fixing failed components and peripherals, wiring, networking or system configuration.
2. We also propose, for the sake of a more effective maintenance of the computer clusters, the integrated computer lab be formally placed under the care / responsibility of the computational physics expertise unit.
3. Individual researchers in the unit will contribute to the maintenance cost of the computing facilities via their research grants.

Space requirement
1. All the computing facilities will be occupying the existing space as they are at the present. These include the present computer labs, theory lab, the server room "bilik Delta" in student center, 2nd floor, Physics School building.
2. In addition, we formally propose the email room located between the computer lab and theory lab be allocated exclusively for the computational physics expertise unit as its physical ‘center’.
3. The computer lab be placed under the care / responsibility of proposed expertise unit (as proposed in (2) in Maintenance).

Expected Output / Evaluation Measures
1. Based on the current publication rate, we expect to publish at least 6 international ISI journal papers per year from the computational physics expertise unit. Even more publication can be generated if the team can have a new academic staff or a post-doc.
2. We will publish at least two joined papers with our experimentalist colleagues per year.
3. We expect to graduate two Ph.D and 2 M.Sc. graduates in three years time.
4. We will establish at least two international research collaborations in the first three years.
5. We will conduct at least one workshop on various computationally-related short courses for the physics school and USM in general annually.

大堀乡赏莲

7月14日周日，日上三竿才起床。颱风过境后的一天，从五楼窗口望出去阳光灿烂，一洗昨天乌沉沉的天色。在房内吃过餐点，趁着还有点时间，开机整理了东能所交代下来的研究计划申请书，添加了两行不知是否真实的内容。两个小时后，货送出去，比之前林老板所订下的时限要晚了一小时多，不知老板会不会不收货。不理了，关了机，到观音乡赏莲去，一偿昨天游厉不成的的些许遗憾。

在宝莲寺下车，走进寺内的舍利塔礼佛，顺便跟秘书处的小姐问明路向。这张观音像就是在宝莲寺照的。

在田园间走了老半天，稻田的景观十分辽阔，在村野的小径中做阡陌行。村名大堀村。时，天微雨，行程很是有味道。

大田莲苑一景，池中有黑白天鹅一对，但不在镜头中。

按图索骥，步行来到大田莲苑隔壁的林家古厝休闲农场，终于可以停歇一番。

这休闲农场景观很悠闲，地方也宽阔，中间有大莲塘，中有小鸭站在莲叶上戏水，可爱。

王莲是农场的卖点。

林家古厝的莲田应该是很美丽的，但是昨天颱风扫过后，莲田的美景都被破坏掉，莲花也被摧残得枯萎了。一眼望去，莲叶连绵盖满湖面，但难以找打盛开的莲花。

尝试找就算那么一朵盛开的莲花，在莲田的阡陌中寻觅了许久，都没有找到惊喜，只有含苞待放的莲花。对莲有一种喜爱，主要是读过宋时周墩颖的《爱莲说》：

出淤泥而不染，濯清漣而不妖，中通外直，不蔓不枝，香遠益清，亭亭靜植， 

可遠觀而不可褻

今次的观音乡赏莲之行都是用走的。一个人在陌生的台湾稻田阡陌中独自用脚走路，是非常独特的经验，之前都很少做这样的事。跟当地的路人互动，按着从seven以八十台币买来的桃园区旅游地图寻找方向，也不知道要去到的地点是否值得一游，有种不能确定的感觉。允许自己走这么一遭，给自己的历练再添加一笔。

晚间七点回到宿舍时累了，睡了个饱，晚间十点四十五分醒来，还没吃饭nie。出得会馆，被雨水洗涤过的空气超超清爽，在虫声唧唧中推了一辆爱心脚踏车到后门外的seven填肚子。其实想要体验在这间店吃微波食品很久了，这次适逢睡醒饥饿，乘兴要了个素三杯炒饭，五十元台币吧，再要杯美式中型咖啡，25台币。第一次在seven的店里吃哦。时约晚间十一点，附近的学生们进进出出的，店里好不要热闹。悠闲的骑着爱心单车在中大校园中来去宵夜，在雨后清风的夜色中，感觉很宁静潇洒，一个人享受得很。忽然感恩让我这次来中大年修假的赖老师。

这一天，颱风苏力来过

7月12日，真的尝到何为山雨欲来的气势。我这里经历了人身中的第一次。。。颱风。苏莉自7月12日傍晚在中坜刮起十多级颱风，直至隔天下午雨都还在下不停。苏力中颱凌晨登陆台湾，在傍晚时分，台风圈还没扫到台湾岛，满天就已经铅云重重，风势呼呼不停。乘天还没降雨，在运动场跑了四十五分钟。平时多人的运动场，冷冷清清。晚间狂风开始怒号，隔着本来隔音效果相当良好的窗户夹着急雨敲打，碰碰作响。当夜睡得不好。

7月13日上午，狂风还是没停。但过了午后，雨势减小。出门一看，真的满目蒼夷，满地都是倒树，放在路旁的机车脚车东歪西倒。那天槟岛强风吹过，的确是造成一些破坏，但跟颱风一比，却又不成意思了。

颱风听说早上已经出海，下午天气应该允许我出游了吧？摇了两个电话给桃园及中坜客运，确定公车已恢复川行。午饭后便跳上客运前往观音乡看莲去。观音乡离中大很近，才二十分路程不到。上车时雨势已经缓和，以为可以趁今天空档游观音的莲花坊，哪知到站后，只来得及拍下这张甘泉寺的照片，雨就下个不停鸟。那还没关系，最大的问题是，风太大，从槟城带来的伞太弱，完全不能撑开，不然一定断掉。勉强冒雨兜了街市一圈，台风天，狂风骤鱼的街道在大白天空无一人，冷清都有点过分。吃了个什锦面，结果乜莲都没有看到，就被逼打道回府了。来观音的目的没达到，好似是在空转，但这过程开启着的是我的一个新的生活层次。在大风大雨中，在一个陌生的国度里，独个儿上公车来来回回，箇中点滴在心头。

新竹的空山新雨

在来台后的第一个周末，7月6、7日，是开始跳出框框，进行台湾走透透的计划的时候了。其实个人并不怎么热衷到处爬爬走。背包旅行，只是偶尔为之的事。通常都是有伴，或有人愿意当带头大哥/大姐，全程策划，我才随行。这次尝试朝抵抗力最强的路径走，挑战自己能否换个思维模式，主动都去新地方探索外在的世界之美，跟美丽的人、景、文化互动交流。这中种生活的乐趣，若能内化成default的常态，当然是美事。所以计划中开始的第一个地点，就是新竹。从中坜出发，只35分钟火车的路程，单程只要票价85元台币。若坐区间车，可能更便宜。新竹罗神父的修会是我落脚之处，住上三个晚上的地方。

修会在新竹市东南街与南大路交接处某巷弄内。进来的小巷路宽狭窄，但一转入修会，在空间上有豁然开达的气象。修会内广场有近十多个篮球场大小，附近邻里们都会在树荫下纳凉闲聊。广场中有一亭俸有圣母象，纯白庄严。亭后有座三层建筑物，我就借宿在第三层的第一号房内，冷气房、洗澡间，无线上网配备，一应俱全。托罗神父的福，我可幸福哦。

又是托罗神父的福，修会有机车二辆，有没有路照我可不晓得，但乘二人上天下海基本上不成问题。这是我在新竹三天的坐骑。周六日头够晒，早上八点半左右罗神父的职务稍毕，两人跳上机车兜游新竹市。在市区骑机车很挑战哦，但主要的不是车多的问题。市区的车速，包括机车的车速都很慢。不轻易按喇叭，而且大家都处处相让，那是因为基本上每个司机都多少会违规驾驶，所以大家都遵守相让的默契，路面的安全就此获得很多人为的保障。这里的骑士都懂得珍惜自己的性命（这不也是人之常情吗？）。反观在槟成骑“魔多”机车的草根市民，横冲直撞，在bumper to bumper塞着车的窄路上zig-zag式的穿来插去，使得路面上人人自危。司机们都很没有人性，耐心又差，不顾旁人死活，也不顾自己死活。大马的路上，人命贱价，不值分文。

罗神父带我去的第一个景点是孔庙，侍俸至圣先师的庙宇。孔庙在中华地区盛行，建筑格局基本都相差不大。庙旁还有两处新竹的观光点，动物园和玻璃博物馆。我们照了两张照片，逛了两圈就离开了。罗神父有事得先回修会，只我一人独骑机车继续兜游。

市区内的庙宇，新陈交错，几乎五步一庙，十步一宇。竹莲寺就在修会一箭之遥，是我首先见到的古庙。庙宇内香火一如市区内的其他庙宇旺盛。

后来晚上跟罗神父晚餐后，停在一家便利店买咖啡零嘴时，正逢竹莲寺就在眼前。但见灯光辉煌，怎不能再照多一张？日夜二景参照，视觉效果各有千秋。

兜着兜着，来到只隔一条街的新竹公有菜市场，是有冷气的两层楼的菜市兼卖场，是市民生活的场景。

市场内卖的食品我都很有兴趣，但肚子只有一个，所以也没买太多。图中所卖的是鹅肉，生意好得很。台湾人吃肉最为流行，我虽不持素，但也不愿把太多肉往嘴里送，阿弥陀佛。

摊位上也有很多试吃的小菜，满意的可以打包带走。用料腌过的豆干、矮瓜、蛇豆、银鱼（即我们的江鱼仔，但体型较大）。

还有这鱼柳，腌在酱油加麻油的酱汁内，真的香味四溢，我差点就冲动要买，但因得配饭吃，最后作罢。

其实新竹很幸福，市区虽也车水马龙，可没比得上台北般喧哗扰嚷。在步行都可到达的距离内，即有空旷山区可做森林浴，最典型者莫过于十八尖山。若从修会所在的东南路骑机车，不到几分钟即可到达山脚。我一人傍晚时分冒着雨靠着男人的直觉，骑机车兜了一大圈摸到十八尖山的其中一个入口。

哇，雨中的十八尖山朦朦胧胧，云雾飘绕。因时近傍晚下雨，山客甚少。我独个儿撑伞雨中漫步，心里响起骆炎聪中六是在我的盗版物理考古题复印册上写下的一首歌词，齐豫的《走在雨中》：

當我走在淒清的路上　天空正飄著濛濛細雨

在這寂寞黯淡的暮色裡　想起我們相別在雨中

不禁悲從心中生

（其实我可没有什么悲啦，只是假假扮文艺而已。=b ）

山的走道绕成一个大圈，走完全程要两三公里。路段上多有梯级把景点贯起来，指示牌也非常贴心，即便我从来没来过也能辩清路向。就这点，槟城的山林步道已经完全没得比。

雨势不久停歇，西阳将末，从山上俯视新竹市，四周寂然。罗神父频频提及十八尖山之美，这时算是体解了。

刻意在山上等待夜色的降临，好让空山新雨后的寂冷路灯照明我的路向。这是一个人的浪漫喔。