Mega Science 2 0: Electrical Electronic Sector

MEGA SCIENCE 2.0 Electrical & Electronics Sector Driven by demand for wireless technology, gallium MMICs (power amplifiers, switches etc.) are initiallyarsenide (GaAs) has been the most-used compound comprised of the main market for GaAs wafers and willsemiconductor material in terms of volume. Moreover, continue to fuel the business in the coming years duesapphire and silicon carbide (SiC) have benefited from to the development of sophisticated smartphones, thethe booming LED market, while bulk gallium nitride development of 3G/4G networks, and the increased(GaN) has become essential for blue laser diode makers demand for data communications. Added to that,[Yole Développement] recently, the development of new GaAs-based devices The GaAs substrate market is expected to increase are enhancing the market with associated high-volumeat a compound annual growth rate (CAGR) of nearly applications such as LEDs, which are currently booming11% to more than $650m by 2017, as shown in Figure due to their advantages over traditional light sources.1.7. This market expansion is fuelled primarily by rising Furthermore, other devices, such as solar cells for high-GaAs content in handsets and rising penetration of concentration photovoltaics (HCPV), will also increaseLEDs in general lighting and automotive applications. the development of the GaAs substrate market, but to aFigure 1.6 Percentage of compound semiconductor substrate compared to siliconFigure 1.7 GaAs Wafer Market Trend 11

MEGA SCIENCE 2.0 Electrical & Electronics Sectorlesser degree (Yole Développement). Figure 1.8 shows LEDs are the largest market for sapphire substrate.applications and markets for GaAs devices. In LED market, the major end applications are general Sapphire has emerged as a versatile material useful lighting and LCD backlighting. Other noteworthy LEDto a range of industries in many applications including market include signage and automotive. However, theLEDs, optics, and RFICs. Sapphire has a number advantages of sapphire vary by application. For example,of attributes that make it suitable for a wide range of lighting and industrial and automotive applicationsend markets, including its hardness and resistance to benefit more from long life compared to mobile handsetsphysical damage and chemical erosion. and notebooks, which prioritize high brightness and low power consumption. Figure 2.9 shows the demand forecast of the sapphire ingot for LED application. Figure 1.8 GaAs markets and applications Figure 1.9 sapphire ingot demand forecast 12

MEGA SCIENCE 2.0 Electrical & Electronics Sector Driven by the booming business in white LEDs for substrate. Table 1.4 shows some of the substrates usedLCD backlights and general illumination, sapphire for GaN as well as their characteristics.was expected to become the leading compoundDensity g/cm Linear Coeffcient of Thermal Expansion Table 1.4 SubDsternasteityMaterLiainlseCaroCmopeaffriicsioennt of*10 semiconductor substrate material in terms of g/cm3 Thermal Expansionvolume by 2011. This exceeds 50% of the total of totalcompound semiconductor processed surface area [Yole GaN 6.1 *10-6/K°Développement, Report on Compound Semiconductor 5.59Substrates 2010]. The market for gallium nitride (GaN) devices include Diamond 3.52 1.18two primary sectors, namely LEDs, in which GaNenables much higher efficiency in green, blue and SiC 3.22 10.3ultraviolet applications, and communication devices, inwhich power and linearity are most important. GaN has Silicon 2.6 2.33been used in LEDs for some years, but high-volume.application is just beginning. Two of the most important Native bulk GaN has emerged as an alternative toapplications are in mobile telephone base stations sapphire or silicon, allowing further improvement of LEDand military radar systems. The military appreciates performance. Despite potential performance benefits forthe important part that GaN may play in X-band radar ultra-high brightness (UHB) LEDs, massive adoption ofsystems. This is a high-end application area that GaN GaN wafers remains hypothetical. Taking into accountshould excel in. the historical price reductions of bulk GaN substrates, Although the need for GaN devices has spurred a base scenario outlines where the GaN on GaNresearch and development, the vol. of GaN devices LEDs will be limited only to niche markets. If the GaNbeing sold into all markets is expected to amount to no industry succeeds in replying to the cost pressure frommore than a small fraction of the vol. of GaAs devices LED makers and the price of 4 inch GaN wafers fallssold within the next few years. While GaN is intriguing, below the breakeven price, a more significant adoptionmany more mobile phone handsets and satellite is forecasted. [Yole Développement]. Combining alltelevision receivers use GaAs chips. One active area applications, the demand for 2” GaN substrates will beof research and development into GaN is the matter of more than two times higher in an aggressive scenarionative substrates for GaN. At least six different substrate than in the base scenario. In the best case, the demandmaterials are currently used experimentally (and often will remain relatively stable until 2020.commercially), for GaN. The reason for such a diversityof substrates is that there no GaN ingots exist such as In MMIC applications, Bulk GaN substrates presentthose of silicon. a very challenging scenario. The GaN power device Currently, the native substrates in use for GaN, for industry (MMIC) generated less than USD2.5M inboth LED and high-frequency, high power applications revenues in 2012. However, overall GaN activityare bulk GaN, diamond, silicon carbide, sapphire, has generated extra revenues as R&D contracts,aluminium nitride, and silicon. The most important qualification tests, and sampling for qualified customersconsiderations for the selection of the substrate type for was extremely buoyant. 16 out of 20 established powera given application include the degree of lattice mismatch electronics companies are involved in or will be involvedbetween the active GaN layer and the substrate, the in the GaN power industry.thermal conductivity of the substrate, the difference incoefficient of thermal expansion between the substrate Among the numerous substrates proposed for GaNand the epitaxial layer, and the overall cost of the power devices, bulk GaN solution is definitely beneficial to the device performance. However, Yole Développement remains quite pessimistic that bulk GaN could widely penetrate the power electronics segment unless 4” bulk GaN wafers can be in the USD1,500 range by 2020. The main reason is that, GaN power devices are positioned as a cost-effective solution, between incumbent Silicon 13

MEGA SCIENCE 2.0 Electrical & Electronics Sectorand the ramping-up SiC technologies. If that $1,500 Rubicon Technology based in USA is one of thecost cannot be reached, then no bulk GaN substrate will big players in sapphire substrate that is engaged inpenetrate the market (Yole Développement). developing, manufacturing and selling monocrystalline1.3 GLOBAL PLAYER – A CASE STUDY sapphire and other crystalline products for light-emitting1.3.1 SUBSTRATE MANUFACTURER diodes (LEDs), radio frequency integrated circuits (RFICs), blue laser diodes, optoelectronics and otherSapphire is one of the popular substrates for LEDs and optical applications. The company applies its proprietaryMMICs. There are more than 130 companies working/are crystal growth technology to produce very high-qualityinvolved in sapphire substrate. About 50% of these are sapphire in a form allowing for vol. production of variousfrom China. However, only 10 companies have captured sizes and orientations of substrates and windows. Themore than 70% market shares. This means that there Company is actively developing larger diameter productsare only a few big players in sapphire substrate. Figure to support next-generation LED, RFIC and optical2.10 shows countries that involved in sapphire wafers window applications. In 2010, Rubicon Technologymanufacturing. opened a new State of the art polishing factory in Malaysia. Previously, polishing had been deemed as a back-end process in the substrate manufacturing value chain. All front-end processes in sapphire substrate manufacturing by Rubicon Technology are done in USA. Companies Involved in Sapphire Wafers : Material Finishing - PSS Established & Projects* / Currently qualified for LED Korea Eastern 31 / 8 + 6 PSS Europe foundriesUS 7 / 3 China4/3 65+ / 15+ Japan 7/4 Taiwan 22 / 9 + 8 PSS foundriesIncludes 6 PSS foundries Figure 1.10 Sapphire substrate manufacturing players based on country 14

MEGA SCIENCE 2.0 Electrical & Electronics Sector For another widely used substrate, GaN, 87% of the the market leaders. Japan will continue to dominate themarket is heavily contested by Japanese companies. Bulk/FS GaN market for the coming years. Table 1.5Currently, non-Japanese players are in the small vol. indicates some of the players from different countries forproduction or R&D stages; which is too early to challenge GaN substrates. Table 1.5 GAN substrate global players Japan China US EuropePlayers Sumitomo Sino Nitride UCSB Ammono Hitachi Metals Semiconductor Soraa SaintGobain/ Mitsubishi Chemical Kyma NGK Insulators Nanowin Sixpoint Materials Lumilog PAM Xiamen Goldneye Unipress Furukawa Freiberger AETech1.3.2 LED BUSINESS PLAYER Netherland and Germany while the back-end operationsThree of the biggest players in LED manufacturing are located in various part of Asia. Figure 1.11 to 1.13are CREE, Lumileds and OSRAM. All of their higher show the location of various operation stages for the threevalue chain or front-end operations are located at USA, big players in LED manufacturing (Yole Developpement). Figure 1.11 CREE LED manufacturing facilities 15

MEGA SCIENCE 2.0 Electrical & Electronics Sector Figure 1.12 Lumileds LED manufacturing facilities Figure 1.13 Osram LED manufacturing facilities 16

MEGA SCIENCE 2.0 Electrical & Electronics Sector1.3.3 MONOLITHIC MICROWAVE INTEGRATED were not forthcoming, several of these companies CIRCUIT (MMIC) INDUSTRY PLAYERS regrouped and underwent consolidation. There are still more than 35 foundries of GaAs MMICs and the numberSetting up a MMIC fabrication facility or foundry for of companies for silicon RFICs and MMICs are growingmass production is prohibitively expensive (Robertson, rapidly.ID&Lucyszyn, S). At one time, most major microwavecompanies had their own facilities. When the profits Table 1.6 List of MMIC Foundries Foundry Location Capabilities Processes offered MarketCree Durham, NC, 3” SiC & GaN GaN HEMT & SiC Mesfet Offers SiC foundry services & has aFiltronics product lineCompound USA 6” GaAs 0.2μm pHEMT Offers foundry services & has aSemiconductors Santa Clara, CA, product lineGGSIBM USA 6” GaAs 0.5μm pHEMT, InGap & HBT Only offer foundry servicesKnowledge ON Torrance, CA, Silicon 0.18μm, 0.25μm, 0.35μm, Only offer foundry servicesM/A ON USA 0.5μm, SiGe BiCMOS & 0.13μm,Nitronex Burlington, VT, 0.18μm, 0.25μm RFCMOSSiGe USASemiconductor 6” GaAs InGaP HBT Offers foundry services & has aTranscom Iksan, S. Korea product lineTriquintSemiconductor Lowell, MA & 4” GaAs 0.18μm, 0.5μm & 1μm, Offers foundry services & has a Roanoke, VA, pHEMT, 0.5μm & 0.35μm, Mesfet product line MSAG USA Raleigh, NC, GaN on 4” GaN HEMT Offers new foundry services & has a Silicon product line USA Ottawa, Ontario, SiGe SiGe BiCMOS Offers foundry services & has a product line Canada Tainan, Taiwan 6” GaAs 0.15μm, 0.25μm pHEMT and Offers foundry services & has a 2μm HBT product line Portland, OR, & Dallas, TX, USA 6” GaAs & GaN 0.25μm, 0.5μm & 13μm, 0.25μm, Offers foundry services & has a 0.35μm, 0.5μm,pHEMT, 0.5μm & product line 0.6μm Mesfet, 0.5μm HFET, 3μm InGaP HBTUnited Monolothic Ulm, Germany 4” GaAs 0.15μm, 0.5μm pHEMT and HBT Offers foundry services & has aSemiconductor Orsay, France product lineWIN Semiconductor Tao Yuan Shien, 6” GaAs 0.15μm, 0.5μm pHEMT and Only foundry services Taiwan 1μm, 2μm HBT 17

MEGA SCIENCE 2.0 Electrical & Electronics Sector MMIC manufacturing services are offered by many can vary from $40,000 to $100,000 depending onfoundries in the USA, Europe and the Far East. A process type, number of wafers requested, and anyfoundry process is typically defined by both the type post fabrication steps requested, such as on-waferof active device and the device’s smallest dimensional testing or pick-and-place. The cost typically includesfeature such as gate length in FET devices or the a complete set of masks (usually 8 to 15 layers)emitter width in bipolar devices. Table 2.6 lists most of and 2 to 6 diced wafers ready for pick and place.the commercially available MMIC foundries worldwide.Some other foundries which are not listed in the table are 1.4 MALAYSIA’S POSITION IN COMPOUND either R&D facilities or offer manufacturing capabilities SEMICONDUCTOR INDUSTRYonly to internal corporate customers or for their own lineof products. Only Gallium Arsenide (GaAs) and Silicon 1.4.1 SUBSTRATE MANUFACTURINGfoundries are available for large scale MMIC production,while other semiconductors are used in research labs Substrate manufacturing is a capital-intensive industryor in small foundries which are primarily working on that requires a large budget and high-skilled humanresearch and advancing the State-of-the-art. resources. The worldwide semiconductor market (not When MMIC technology was developed through specific to compound semiconductors) is currentlygovernment funded research from the mid-1980s to valued at RM812 billion (2009) and is expected to grow atthe mid-1990s, MMIC manufacturers maintained small about 7% per year. The output value of semiconductorsfacilities. However, today the most successful foundries in Malaysia is around RM39 billion (2009), a global shareare much larger. It has been estimated that a new MMIC of 5%. However, a majority of the semiconductor marketfoundry requires investment of at least $200 - $500 in Malaysia is predominantly focused in Silicon-basedmillion. In selecting a foundry process, the cost per mm2 products, and the semiconductor firms operating inof wafer is the driving factor to remaining competitive. Malaysia concentrate primarily on assembly and testing. For this reason many foundries are upgrading their Furthermore, while there are several foreign companiesequipment to 6” GaAs wafers to lower the cost to their operating in Malaysia for silicon wafer manufacturing,customers. The cost of a MMIC chip is very low and such as Sumitomo and Shin-Etsu, this is not the caseranges from $0.20 to $3.0 per mm2 depending on the for compound semiconductors.size of the chip and the process used in manufacturing. Malaysia’s role in front-end substrate manufacturingSiGe MMICs are the least expensive, while 0.15µm (other than silicon) either through foreign or localpHEMT, GaN HEMT and SiC Mesfet are the most companies is still non-existent. In 2010, Rubiconexpensive. Nevertheless, MMIC development is technology opened a State-of-the-art facility responsiblecostly and time consuming; the cost of a foundry run for the labour-intensive crystal polishing process for Figure 1.14 Substrate manufacturing value chain 18

MEGA SCIENCE 2.0 Electrical & Electronics Sectorwafer production. Polishing is a back-end process in carbon economy and achieve sustainable developmentthe substrate manufacturing value chain. Figure 1.14 after the worldwide energy crisis and the global financialshows the value chain for substrate manufacturing. meltdown of 2007-2008. Through this policy, a Green1.4.2 LIGHT EMITTING DIODE (LED) Technology Financing Scheme (GTFS), amounting to about USD494.97 million (RM1.5 billion) was established that was available to manufacturers of green products and developers using green technology.IN terms of LED products, Malaysia currently exports These government efforts have supported andRM 1.8 billion (2008 data) of LEDs, which represents sustained the continuous growth of Malaysia’s LEDapproximately 10% of the global LED market, with most industry. In addition, Malaysia ceased all production,of these exports in the illumination (luminaire) sector. import, and sale of incandescent light bulbs bySolid State lighting is one of the fastest growing sub- January 2014 as part of an efforts to save powersegments, projected to grow at 28% per annum over and cut greenhouse-gas emissions. Currently, fewerthe next 10 years, to a size of RM170 billion (Economic than 5% of lights in Malaysia are LEDs, providingTransformation Programme – A roadmap for Malaysia) LED companies with an untapped domestic market. Currently, four of the largest global SSL companies 1.5.2 ECONOMY TRANSFORMATIONeither operate in or control a significant portion of PROGRAMME (ETP)companies in Malaysia. There are also a number ofcompanies involved in other areas of SSLsuch as contract Another government initiative, the Economicmanufacturing, systems integration, LED packaging, Transformation Programme (ETP), aims to utiliseand application design, as well as speciality companies Malaysia’s economic sectors to propel the countryfocused on various heating and optical elements. into a high-income nation. The programme details 15 Entry Point Projects (EPPs); specific to Malaysia’s1.4.3 MONOLITHIC MICROWAVE INTEGRATED semiconductor, solar, LED, industrial electronics, and CIRCUIT (MMIC) home appliances sectors. The EPP for the LED sector focuses on the following:Setting up a MMIC fabrication facility for mass • To attract more substrate and epitaxy manufacturersproduction is an expensive investment prospect. Otherthan expensive fabrication equipment, another major to complete the solid-State lighting (SSL) valueinvestment required is in CAD facilities. The task of chain in the country;designing a competitively priced circuit should not be • To build more LED wafer fabrication plants to furtherunderestimated and extensive CAD facilities along with develop the country’s LED cluster; andexperienced designers are essential. While there are • To aggressively cultivate Malaysia’s strong positionseveral small scale R&D based compound semiconductor as a packaging hub for SSL products.fabrication facilities in Malaysia, there is no commercialcompound semiconductor foundry in Malaysia.1.5 COUNTRY POLICY The programme also plans to develop at least five1.5.1 MALAYSIA’S LED INDUSTRY ROADMAP Malaysian companies into regional and global brands for SSL products and/or components by 2020. The goalIn 2009, the Malaysian government launched the is to have at least one local SSL company become aNational Green Technology Policy to move toward a low globally recognised name in the SSL market each year from 2012 to 2016. 19

MEGA SCIENCE 2.0 Electrical & Electronics Sector1.5.3 THE CHALLENGES replace conventional filament and incandescent-based lights. The reason for this is simple. Solid State LEDsCost is a major issue in the growth of the market. are versatile, with strong integration capabilities due toThe worldwide LED chip market is dominated by big small size and minimal power consumption.companies such as Osram, Nichia Corporation., and Globally, many companies are working and participateCree Inc. Initial investment in this sector is high and in the compound semiconductors. However, significantcountries such as Germany, Japan, and the US are market share is held by big players from Europe andfar ahead technology-wise compared to Malaysia. the United States. This chapter has presented theNevertheless, in Malaysia, a majority of the LED fundamental basis of the industry in terms of theplayers are focused on LED packaging. To produce technology and drivers; hence, it will serve as a goodLED modules, a company must invest in machinery and starting point for ventures in this competitive industry.quality materials. According to Razali Mohammed, CEO of the MyLEDGroup, in order to build LEDs, 70% of the machineryshould be built in-house. “Our automation machinecosts USD2.65-3.31 million (RM8-10 million) forone LED assembly line alone,” Mohammed said.MyLED Group is a local company that has both SSLand LED manufacturing technology. The firm hasinvested UD54.90 million (RM175 million) to expand itsmanufacturing capability in the green technology sector. Another factor that adds to the costs of LEDproduction is the lack of availability of raw and qualityintermediate materials. Most of the materials used toproduce LED modules such as the wafers, ceramicsubstrates and silicon are imported. Although someof the materials may be available locally, a number ofLED packaging firms still import materials in order tomeet quality requirements. To illustrate, Osram is theonly company that has its own wafer fabrication facilityMalaysia. Manufacturers in Malaysia are still dependenton imported chips for producing quality modules.1.6 CONCLUSIONCompound semiconductors are an emerging industrywhich has recently seen tremendous growth. This hasbeen driven by the booming smartphone industry andincreasing demand for LED applications. Currently, LEDdevices are widely used in different type of applicationssuch as mobile phones, display devices, and theautomotive industry, and will continue to grow in thenear future with increase LED lighting application to 20

MEGA SCIENCE 2.0 Electrical & Electronics Sector21

MEGA SCIENCE 2.0 Electrical & Electronics Sector 2 22

MEGA SCIENCE 2.0 Electrical & Electronics Sector2CHAPTER 2COMPOUND SEMICONDUCTOR: MOVING UP THE TECHNOLOGY VALUE CHAINThe Electronics and Electrical sector is an important Malaysia’s focus on E&E, especially in thecontributor to Malaysia’s economy, accounting in 2009 for semiconductor, LED and IC industries, has traditionally6% of Malaysia’s gross national income (GNI), 522,000 been on assembly or other lower value-addedjobs, and 41% of Malaysia’s total exports [ETP, chapter segments of the industry. Meanwhile while countries11]. The E&E industry has also played a major role in the like Taiwan, South Korea and Singapore have captureddevelopment of the Northern Corridor (semiconductors the higher value-added activities in R&D, design andand industrial electronics), Klang Valley (sophisticated manufacturing. Hence, it is crucial for Malaysia to moveservices), Johor (logistics intensive E&E manufacturing) up the value chain while maintaining the current back-and Sarawak (developing cluster for silicon substrate end segments. The idea is to have the most coveragemanufacturing). along the total value chain in a particular industry. Malaysian E&E has a strong foundation in This will not only provide a stronger contribution tosemiconductors and industrial electronics. Moreover, the country’s economy along with the entailed jobthere are some emerging sectors growing globally opportunities, but will also provide an opportunity forat more than 20% a year, such as solar photovoltaic Malaysia to contribute in the emerging technologytechnology (solar) and light emitting diodes (LEDs). by capitalising on its abundant natural resources.Malaysia already enjoys a strong base of companiesand is poised to capture global growth in manufacturing 2.1 MOVING UP THE VALUE CHAIN: CONCEPTUALcapacity. FRAMEWORK According to World Bank report, “Moving up the value chain concerns the process of shifting productive 23

MEGA SCIENCE 2.0 Electrical & Electronics Sectoractivity of a nation, an industry or a firm towards the The electrical and electronic sector is an importantproduction of goods and services that generate higher contributor to Malaysia’s economy, with exports invalue added. While at the surface this might come the range of RM250 billion in 2008. It employs moreacross as a fairly straight-forward process, moving up than half a million people, many in better-paid skilledthe value chain is an inherently complex undertaking. positions. With about 1,900 active companies, theIt requires a fundamental shift in the sources of growth sector has seen many success stories, especially inand competitiveness. To move up the value chain, the fields of semiconductors and industrial electronics.competitiveness can no longer be measured merely in Malaysia is home to many of the largest and mostterms of the volume of goods and services that can be successful companies in the field such as Rubicon,produced at the lowest possible cost. Instead, it needs OSRAM, Lumileds, and Agilent, and has incubatedto be measured by the amount of domestic value added home-grown stars like Silterra and several research andthat can be generated by globally competitive firms development centres for compound semiconductors. Itoperating in Malaysia.” has seen a strong start in the fast-growing subsectors of However, the proposition made in this mega science solar and light emitting diodes.framework, in particular the electrical and electronic As such, there are faced by in several challengessectors, is not entirely in agreement with the World Bank for Malaysia increasing the need to move-up the valuereport; instead, moving value chain should not only chain include (ETP Roadmap) such as the following:be measured by the amount of domestic value added E&E’s contribution to Malaysia’s exports and itsgenerated by global firms operating in Malaysia, but economy has in decline.also the amount of domestic value added generated by E&E’s share of Malaysian exports increased dramaticallylocal Malaysian companies. It is a roadmap for domestic during the 1970s and 1980s as Malaysia industrialisedlocal companies in Malaysia take part in the higher and introduced enabling policies, such as free tradevalue-added business. Although, it is good for the global zones. However, since 2000, E&E exports have grownfirms to bring higher-value chain to Malaysia, greater more slowly than other exports (at 0.4% for E&E versusemphasis should be placed on setting up local firms 7% for all exports), resulting in a decline in its share ofthat play a bigger role in the front-end of the industries. exports from 59% in 2000 to 41% in 2009. Malaysia’s E&E sector is facing increasing2.1.1 THE NEED AND URGENCY TO MOVE-UP THE competition. VALUE CHAIN China, which has emerged as the world’s factory, is a significant threat. A World Bank study shows theMoving up the value chain is central to Malaysia‘s increase in export competition between Malaysia andaspiration of joining the league of high-income China [World Bank 2009]. In 2007, 59% of Malaysia’seconomies. Malaysia seems stuck in a middle-income exports to EU were under threat from China, comparedtrap, a predicament which prevents middle-income to only about 31% in 1990. Other emerging Asiancountries from taking the next step on the development economies such as Vietnam are fast developing low-path towards high income. This has manifested itself in cost companies in the E&E industry; while at the high-the growing inability to remain competitive as a high- end, Singapore and Taiwan compete for investments invol., low-cost produce coupled with the difficulty to higher-value added activity.break into fast-growing markets for knowledge-andinnovation-based products and services. The implicationis that, despite past growth successes, living standardsmeasured by per capita gross national income couldhave been significantly higher (World Bank report). 24

MEGA SCIENCE 2.0 Electrical & Electronics SectorConcentration of activity in assembly results in Monolithic Microwave Integrated Circuit (MMIC) willlower value-added gains. be explored, and opportunities that can be taken byWhile Malaysia has built up significant clusters in E&E, Malaysia as well as their economy contribution will bemuch of the activity is in relatively low value-added proposed.assembly, rather than higher value-added activities such 2.2.1 SUBSTRATE MANUFACTURINGas component manufacturing or R&D. For instance, for The substrate manufacturing industry involves severalPenang, with its semiconductor cluster, most of the capital intensive sectors that focus on growingactivities are on assembly and testing rather than higher semiconductor material that can be made as the wafervalue wafer fabrication. As a result, the value added or substrate for compound semiconductor devices. Theper worker of about RM70 thousand is comparable value chain starts with crystal growth activity of theto that of China and only a fifth of Singapore’s. said semiconductor material. Once the crystal is ready, several activities or processing are carried out, including2.2 MALAYSIA’S STANDING IN THE GLOBAL drilling, slicing, grinding, polishing and cleaning. COMPOUND SEMICONDUCTOR INDUSTRIES Figure 2.1 depicts the value chain of the substrate manufacturing industry.In this section, the value chain for substratemanufacturing, Light Emitting Diode (LED) and Figure 2.1 Substrate manufacturing value chainAluminium oxide is heated in a As the furnace cools, the sapphire 2-inch to 8-inch cores are drilledfurnace to 3,727 ℉ crystal-known as a “boule”- is born from the boulesCores are sliced into waters, Boules typically The wafers are diced into small chipsand the wafers are polished range in size from which are assembled into an LED module 3-100kg Epitaxial Wafers go through epitaxial growth Layers process to deposit very thin layers of Sapphire Substrate material to build the LEDSource: Rubicon Technology Figure 2.2 Sapphire substrate manufacturing 25

MEGA SCIENCE 2.0 Electrical & Electronics SectorFor instance, Figure 2.2 illustrates all the processing teardrop shaped ‘boule’ of sapphire material (step 2 insteps involve in sapphire substrate manufacturing, per Figure 2.2). Once the single crystal sapphire boules arean infographic from Rubicon Technology. In this Figure, grown they are cored-drilled into cylindrical pieces (stepthe process steps for the substrate/wafer making are 3). Wafers are then sliced from these cylindrical coresuntil step 4. Steps 5 and 6 are the processes involve in (step 4).LED device manufacturing. Currently, Malaysia’s status in this value chain,2.2.1.1 OPPORTUNITIES especially related to sapphire substrate manufacturing, is at the lower-end (polishing and cleaning). One ofThe raw material for sapphire substrate is aluminium the global players thathave established a plant foroxide (Al2O3). Aluminum oxide is produced industrially this purpose is Rubicon Technology. The front-part isfrom the mineral bauxite. Bauxite deposits worldwide handled by the USA headquarters and the lower-endare estimated at approximately 20 billion tons, while is by Rubicon Malaysia. Looking at the value chain forannual worldwide output amounts to about 100 million substrate manufacturing, at least two different localtons. Australia has the largest output and deposit levels. companies can be created to cover the complete valueAluminum and aluminum oxide are manufactured chain. One company can work on the front end (crystalusing the Bayer method. Bauxite is crushed, dried, growing and ingot/boule production) and anotherand dissolved using concentrated sodium hydroxide company on the lower-end (drilling, slicing, and wafersolution. Impurities such as iron, silicon, and titanium polishing)are separated from the bauxite in so-called red mud. Malaysia produces bauxite mineral (British GeologicalAluminum hydroxide is precipitated from the solution Survey). In 2007, Malaysia produced 156,785 tonnes ofand calcinated at 1200-1300 °C to form Al2O3. bauxite; by 2009 the amount was up to 280 thousand Fine aluminium oxide powder or commonly known as tonnes, as shown in Figure 2.3. This corresponds toalumina powder is added to an oxyhydrogen flame and less than 1% of the worldwide production. Table 2.1this is directed downward against a mantle (Heaton). indicates the amount of bauxite produced yearly by theThe alumina in the flame is slowly deposited, creating a country between 2003 and 2009. Figure 2.3 Malaysia yearly bauxite productionSource: United States Geological Survey (USGS) Minerals Resources Programme 26

MEGA SCIENCE 2.0 Electrical & Electronics Sector Table 2.1 Malaysia Yearly Bauxite ProductionYear Production Unit of Measure % Change2003 5.69999980926514 Thousand metric tons NA2004 2.03999996185303 Thousand metric tons -64.21%2005 4.7350001335144 Thousand metric tons 132.11%2006 91.8059997558594 Thousand metric tons 1,838.88%2007 156.785003662109 Thousand metric tons 70.78%2008 295.175994873047 S Sy Thousand metric tons 88.27%2009 280 Thousand metric tons 5.14% Malaysia’s bauxite resources are located at Bukit Batu, Labuk valley in the State of Sabah. The two operatingBukit Gebong, Lundu-semantan and Tanjung Seberang bauxite mines in Malaysia are located at Sungai Rengitin the State of Sarawak and at Bukit Mengkabau and in the State of Johor. 100 Bauxite output in 2006 shown as a 10 percentage of the top producer 1 (Australia - 59,969,000 tonners) Figure 2.4 Worldwide bauxite output 27

MEGA SCIENCE 2.0 Electrical & Electronics Sector2.2.1.2 RECOMMENDATIONS challenging industry, it is important to be part of the value chain, working towards better quality substrate materialEven though the amount of bauxite resources in Malaysia through R&D and at the same time able to produceis small compared to worldwide production, this may be cheaper wafer for example, a 6 inch sapphire waferthe starting point for Malaysia to venture into front-end that cost less than USD4. The market opportunity forsapphire substrate manufacturing. Once the focus has substrate industries such as sapphire or other compoundbeen set, further exploration can be made to find the raw semiconductor wafers is at the large wafer level. Basedmaterials. In addition, the raw material can be imported on data extrapolation, the market opportunity for 6-inchfrom neighbouring countries such as Indonesia, which sapphire wafer will continue to grow until 2020. Figurehas 16,000,000 tonnes of bauxite if the demand is high. 3.6 shows the market opportunity for sapphire wafer.Figure 3.4 shows world bauxite output in 2004. Other than sapphire, as a substrate manufacturer One of the key advantages of moving-up value chain there are several materials that can be investigated toin substrate manufacturing and setting up local players be produced locally. Some of the materials that currentlyfor substrate production with local raw materials is the being used as the wafer for compound semiconductorassumption that the price for the wafer can be further devices are Gallium Nitride, Silicon Carbide and Silicon.reduced. This will not only help Malaysia gain significant The following is a list of advantages and disadvantagesmarket share, but will also contribute to bringing down of wafer materials (Widney, Douggreentechmediathe price of the compound semiconductor-based 2012):devices. Sapphire: Users include almost all players except Figure 2.5 shows the forecast for sapphire wafer price Cree Lattice mismatch: 13%;until 2020. To remain competitive and successful in the Figure 2.5 Sapphire wafer price forecast Figure 2.6 Market opportunities for Sapphire waferSource: Canaccord estimate model Source: Canaccord estimate model 28

MEGA SCIENCE 2.0 Electrical & Electronics SectorAdvantages: Stable, mismatch well researched; 2.2.2 LIGHT EMITTING DIODE (LED)Disadvantages: Quite expensive.Silicon Carbide: Users: Cree; Lattice mismatch: 3.5%; The Light Emitting Diode (LED) value chain consists ofAdvantages: Very stable. Low mismatch aided several steps to manufacture LED devices on top of theby cancelling thermal mismatch, highest thermal wafer/substrate. The value chain can be divided intoconductivity; two section based on the value-added: front-end or highDisadvantages: Priciest, almost proprietary; value-added and back-end or low-value added.Silicon: Research maturing in OsramOpto, Bridgelux, 2.2.2.1 OPPORTUNITIESChina; Lattice mismatch: 17% (plus a 56% additivethermal mismatch); The front-end part starts from a substrate/wafer. TheAdvantages: 80% substrate cost reduction potential, substrate may come from a different industry (e.g.ubiquitous, big wafers; substrate manufacturing industry). The next step willDisadvantages: Lattice and thermal expansion huge be the semiconductor material epitaxial process on themismatches; wafer. The type of semiconductor material depends onGallium Nitride: Users: Soraa; Lattice mismatch: 0%; the colour of the LED. Different materials with differentAdvantages: Very good Lattice and thermal match, energy band-gap will produce different colour. Tohomogeneous material allows higher drive level by illustrate, Gallium Nitride (GaN) is used for blue LED.tuning the GaN for reduced droop; and Once the epi-wafer has been completed, the deviceDisadvantages: Unstable, defect-ridden and quite fabrication process begins. This includes lithography,expensive. etching, metallisation and several other fabrication Another opportunity is venturing into totally different processes. At the end of the process, the LED dyes willmaterials, which requires huge effort in terms of research be created on the wafer. This marks the end of the front-and development and a long time before gaining any end part.market share. This is suitable for long term target. Some The back-end activity starts from the dyes on the wafer.of the materials that can be researched include glass, The wafer with fabricated LED will be diced into individualgermanium, and Aluminium Nitride (AlN). LED dyes. This individual LED dye will be binned based on the quality and packaged with other circuitry and housing. For a white LED, a phosphor coating will be applied so that the white colour can be produced when the blue light hits the phosphor. The packaged LED can be assembled into different types of LED lamps of lights creating solid-state lighting. Figure 2.7 illustrates the complete value chain for blue LED manufacturing. The basic steps will be the same for different colour LEDs, except for the semiconductor materials and some extra processes like phosphor coating. 29

MEGA SCIENCE 2.0 Electrical & Electronics Sector Figure 2.7 LED device manufacturing value chain The production of LED lamps and luminaires can companies as well as global companies that outsourcebe termed LED lighting or solid-State lighting (SSL). their manufacturing to Asian subcontractors such asIf LED lighting is included with LED chip fabrication, Cree, Philips Lumileds and OSRAM. OSRAM and Philipsthe total value chain including all the elements in LED Lumileds maintain packaging and manufacturing plants intechnology can be divided into 5 different segments as Malaysia. Figure 2.8 reveals certain major players in thefollows: materials, LED chip and lighting components, LED industry, along the value chain. It can be seen fromfinal product, distribution, and sales as shown in Figure the Figure that companies such as Osram and Philips2.8. Lumileds play major role in the upstream and downstream The manufacture of solid state lighting comprises of the value chain while some companies like Cree andproduct design, product manufacturing, marketing Nichia only focus on the higher value added stage.and selling. This is the lower-value added part of thevalue chain and many Asian countries take part inthese activities, including Malaysia. This includes local 30

MEGA SCIENCE 2.0 Electrical & Electronics Sector Figure 2.8 Certain big players along the LED value chainSource: A.T. Kearney Analysis2.2.2.2 RECOMMENDATION Currently, four of the largest global SSL companiesGenerally, Malaysia already participates in the back- either operate in or control a significant portion of workend or lower-added in the value chain. Malaysia to companies in Malaysia. There are also a numbercurrently exports RM1.8 billion (2008 data) of LED of companies involved in other areas of SSL such aswhich represents approximately 10% of the global LED contract manufacturing, systems integration, LEDmarket, with most of these exports in the illumination packaging, and application design as well as speciality(luminaire) sector. Thus, Solid State lighting is one of the companies focused on various heating and opticalfastest growing sub-segments, projected to grow at 28% elements.per annum over the next 10 years to a size of RM170billion (Source: Economic Transformation Programme –A Roadmap for Malaysia). 31

MEGA SCIENCE 2.0 Electrical & Electronics Sector Figure 2.9 LED lighting industry value chainSource: Centre on Globalisation, Governance and Competitiveness, CGGC report Since Malaysia has already established itself in the 2.2.3 MONOLITHIC MICROWAVE INTEGRATEDlower-added value of the value chain, it is the time CIRCUITS (MMICS)for the country to move-up to the higher-value addedsectors and the local companies should play major role The value chain of MMIC manufacturing more or lessin this activity. Climbing the higher-value-added value similar with silicon-based CMOS IC manufacturingchain means Malaysia must venture into LED device where there are two different value chains, one for thefabrication. This involves epitaxial on-wafer processes circuit design process where the whole design processand device fabrication. In other words, Malaysia starts with specification, until layout will be done by arequires a compound semiconductor fabrication facility design house and the second value chain involves theor foundry. This foundry can serve the fabrication of fabrication process of the MMIC. In general, once amonolithic microwave integrated circuit (MMIC) as well. MMIC circuit has been designed, the design will be sentIn this regard, at least two companies can be created to a foundry for the fabrication. The dissimilarity fromone for the epitaxial process and another one for the CMOS fabrication lies at the fabrication process steps,device fabrication. as in MMIC fabrication, compound semiconductor materials will be used and therefore, the process involves epitaxial growth of the compound semiconductor on the 32

MEGA SCIENCE 2.0 Electrical & Electronics Sectorsubstrate. Figure 2.10 shows the simplified value chain LED device fabrication, all of them inter-related andfor MMIC fabrication. complement each other (i.e. substrate manufacturing2.2.3.1 OPPORTUNITIES industry requiring the MMIC or LED fabrication industrySimilar to LED device manufacturing, the process starts and vice versa).with an appropriate substrate or wafer. As explained Once the substrate is ready, the next process isearlier in this chapter, substrate or wafer alone is an epitaxial growth of the compound semiconductorindustry by itself and has its own value chain. Therefore material on the wafer. Next, is the device fabricationsubstrate manufacturing and MMIC fabrication or on the epi-wafer that it use to complete the integrated circuit. (The individual dies on the wafer will be sliced and packaged, then be integrated on the MMIC system).Figure 2.10 MMIC fabrication value chain2.2.3.2 RECOMMENDATION in Malaysia concentrate primarily on assembly andMost integrated circuit designs and fabrication activities testing. However, there are few a few CMOS IC designin Malaysia focus only on the CMOS design process, house in Malaysia and one locally owned CMOS ICwhich involves semiconductors other than compound foundry (Silterra).material. The worldwide semiconductor market (not Currently, there is no commercially viable localspecific to compound semiconductor) is currently valued company for MMIC design and no local player for MMICat RM812 billion (2009) and is expected to grow at about fabrication. Nevertheless, there are several research7% per year. The output value of semiconductors in and development centres located at the universitiesMalaysia is around RM39 billion (2009), a global share and government-linked companies conducting researchof 5%. However, the majority of the semiconductor on MMICs. However, due to the lacking of compoundmarket in Malaysia is predominantly focused on silicon- semiconductor foundry, the research cannot bebased products and the semiconductor firms operating translated into actual products. 33

MEGA SCIENCE 2.0 Electrical & Electronics Sector Malaysia should take this opportunity to participate inMMIC value chain from front-end all the way to the back-end. Local IC design house can created and one localcompound semiconductor foundry can be developed tocreate the complete ecosystem. Together with substratemanufacturing industry and compound semiconductorfoundry, the whole value chain can be covered, not onlyfor the MMIC industry but also the LED device industry.2.3 CONCLUSIONThe compound semiconductor industry, through lightemitting diode (LED) applications and monolithicmicrowave integrated circuit (MMIC), is a billion dollarsindustry and will continue to grow rapidly. Industries allover the world are working very towards the improvementof the technology, as mention in finding better materialsand techniques to be the market leader and controlsignificant portion of the market share. While the number of companies and countriesparticipate in this industry is huge and reach variouscontinents but significant market share is controlled byfew companies from the US and Europe. These fewcompanies participate in the front-end part of the valuechain. The back-end part is taken by many companiesfrom Asian countries. Malaysia has taken part in this industry and is nowactively participating in the back-end of the value chain.In order to become a big player in the industry, Malaysiamust move up the value chain and become involved inthe higher-value segments. This requires huge effortand investment, be it ineithermonetary, human capital,infrastructure or governmental policy terms. Thischapter has explored the complete value chain relatedto compound semiconductor industry in particular forthree different sectors: substrate manufacturing, LEDmanufacturing and MMIC manufacturing. As such asdetailed understanding of the value chain is requiredbefore necessary action plan can be proposed for thetransformation of the country toward higher-value addedsectors. 34

MEGA SCIENCE 2.0 Electrical & Electronics Sector35

MEGA SCIENCE 2.0 Electrical & Electronics Sector 3 36

CHAPTER 3 MEGA SCIENCE 2.0 Electrical & Electronics Sector3COMPOUND SEMICONDUCTOR: PLAN OF ACTIONS AND ROADMAPThe electronics and electrical sector is an important 3.1 S COPE AND STRUCTURE OF THE ROADMAPcontributor to the national economy, accounting forRM37 billion GNI ,which represent 6% of the nation GNI, The Science, Technology, and Innovation (STI) sector522,000 jobs and 41% or Malaysia total exports in 2009 will be a strategic key player and the Mega Science(Economic Transformation Programme report). Agenda will be the bedrock and platform for the journey. Nevertheless, E&E sector in the country faces For this purpose, E&E has been chosen as one of thesignificant challenges to maintain the growth especially sectors for the Mega Science Framework, and thefrom the competition from China, Taiwan, Singapore compound semiconductor industry has been identifiedand many Asian countries. Under the national as one of the subsectors that have potential for STItransformation programme, E&E sector has been sustainable development and growth.chosen as one of the sectors that can move the country To achieve this ambitious goal, sector members haveforward and achieve the vision 2020. The public, undertaken an effort to develop a series of roadmaps andprivate sectors, and the whole nation should face action plans covering different segments of the sectorthe challenges and take part in the national agenda including research and development (technologicalto achieve the vision and move forward beyond it. roadmap), institutional framework and policies and infrastructure development. This roadmap provides a foundation for identifying a set of effective technologies, economies, policies, institutions, and milestones that will allow compound semiconductor industry in Malaysia 37

MEGA SCIENCE 2.0 Electrical & Electronics Sectorto grow and give a significant contribution to the national This step is different from the conventional roadmapeconomy, technological advancement and human methodology that uses forecasting method. In backcastingcapital development, as well as being part and parcel of the necessary paths are developed starting from thebecoming a developed nation. future (based on scenario) and bringing back to present3.2 ROADMAP METHODOLOGY time. This is to avoid constraining the mind with currentThis roadmap was developed and compiled using limitations hence promoting outside the box kind ofinputs from wide range of stakeholder information thinking. Once the development paths have beenobtained through several methods, including internet completed through backcasting, necessary action planssearch, workshops, surveys and interviews from the can be prepared to fulfil all the milestones along theLED industry, IC industry, wafer/substrate supplier, path.academic institutions and government agencies. This Chapter 1 and 2 of this report describes the baseroadmap should be regarded as a work in progress and line, including the technology and the value chainas the technology, market, and regulatory environments of three different industries related to compoundcontinue to evolve, some information and analyses will semiconductors: substrate manufacturing, LEDs andneed to be updated and additional tasks added. MMIC. The proposed roadmap will cover all these industries as a whole, and emphasis will be given for Malaysia to participate in the total value chain including upstream and downstream. Hence, there will be three different stages of the roadmap; starting with the short term plan (2013-2020), medium term plan (2020-2030) and long term plan (2030-2050). 3.2.1 SCENARIO BU Figure 3.1 Roadmap Methodology The future scenario that will be predicted is at the year The roadmap methodology consists of four different 2050 and beyond which is the long term stage. Thesteps, as depicted in Figure 3.1. The development scenario for Malaysia, by the year 2050, is that it willstarts with defining focus areas. This means to select be a market leader for consumer electronic productsseveral areas under the research theme that will be the with significant market shares. This scenario is madecore sectors. For compound semiconductor sector, the due to the correlation between developed countries withfocus areas are substrate manufacturing, LEDs, and electronic industry.MMIC. The next step is scenario building. Basically this To illustrate, Japan is a developed country and itsis the process of predicting the future before it happen. electronic industries are one of the main driving factorsThe scenario must be something futuristic yet plausible. to this. Companies like SONY started just after theOnce the scenario or scenarios have been defined, the Second World War, when Japan lost a war. And, now,next step will be backcasting. they are one of the main contributors that have shaped Japan to what it is today. Samsung is another electronic industry that has helped Korea to be one of the high income countries. Both of these companies have strong support from their governments as they are seen as valuable national assets. Hence, this is what Malaysia should do to establish our own electronic industry as soon as possible. 38

MEGA SCIENCE 2.0 Electrical & Electronics Sector One of the challenges that will be faced by Malaysian industry market. Nevertheless, smartphones are not theelectronic industry is strong competition from global only business entity that Malaysia will take part in. Someestablished industries. There are several ways to other consumer electronic products may be included.tackle this problem. For a start, Malaysian electronic The proposition is that a Malaysian electronic companycompanies should target to capture the local market may become the next Sony, Samsung, or Apple.instead of going global. This can be done if the product Furthermore, this scenario will also complete theis high quality and State of the art technology as well as eco-system for the product cycle. The companies thatpriced reasonably. produce the next generation smartphone and other Pricing and performance are key factors and there is electronic products will serve as the market for all othera possibility to achieve the target if the complete value industries. In other words, this creates a market for allchain can be done locally. For instance, smartphone other businesses such as wafer/substrate manufacturer,has been chosen as the first product to establish Fabrication Company, packaging company, IC designMalaysian electronic companies. In order for Malaysian house, and sales and distribution. Figure 3.2 shows thissmartphone to attract significant market share locally, scenario that will set the targets for Malaysia to becomethe phone should be competitively priced with State of a key player in the compound semiconductor industry.the art technology or at least on par with other brands. To Looking at the scenario, one of the key aspects is cost.be part of the global market and gain significant market As seen from the above list, low cost is one of the factorsshare, a scenario in which Malaysia is a global player of the scenario. Therefore, covering as much as possibleand market leader can be predicted. This scenario can the whole value chain from front-end to back-end willbe made a possibility with the following targets: be crucial in achieving this target. Advance technologyi. New display technology that includes a translucent through R&D activity will also be part of the low-cost activity. In addition, new institutional frameworks, policy and flexible display. and infrastructure need to be developed to achieve theii. New MMIC technology with low cost, low power, final vision. Another area that can be considered in order to and ultra-wide bandwidth system on chip. establish the Malaysia electronic industry is to beginiii. New substrate technology that offers very high with products or solutions that solve local issues and problems. This approach is similar to how technology quality and low cost for LED and MMIC fabrication is developed by the Americans, as they start with or solutions/products that are required by their military.iv. LED display technology that uses other materials The military requirement opens up opportunities for local for LED device fabrication such as plastic or glass. companies to come up with ideas and solutions, whichAs can be seen from this scenario and the events behind at the end lead to new breakthroughs and overtimeit, that scenario covers all focus areas in the compound these breakthroughs reach consumer electronics.semiconductor industry (e.g. substrate manufacturing, Similarly, the Malaysian Government can initiateLEDs and MMIC) grand challenges ideas/projects crucial to Malaysia. In this scenario, low price and high performance For example, say that the local broadcasting mediasmartphone is chosen as the business entity and the are attempting all-digital broadcasts and that for thesemarket share as the key performance index (KPI) for purposes Malaysians need digital set-top boxes inthe scenario due to its role in driving the compound every house. This can be a grand challenge to local 39

MEGA SCIENCE 2.0 Electrical & Electronics Sector Figure 3.2 The Scenariocompanies/individuals/groups to propose a digital set- 3.2.2 ROADMAP STAGEStop box that is Malaysian made or has at least 70%local contents. This may be the first product to establish There are three different stages for the roadmap: firstMalaysian electronic companies. term (2013-2020), medium term (2020-2030), and long Due to the demands for this product, the market size is term (2030-2050). Therefore, there will be three differentsubstantial and every value chain along the production features to cover all the stage and each feature will havewill benefit from the ready market. This product may also several events to describe the feature. The following arebe a catalyst to spearhead bigger business opportunities the features and events for the three different terms:in other segment of the electronic industries. There are Establishing the foundation for the compoundmany other challenges that can be thought of which semiconductor industrysolve Malaysian problems. The government can play a In the near term or for the short term, compoundbigger role to initiate the challenges. semiconductor industry in Malaysia will be established. 40

MEGA SCIENCE 2.0 Electrical & Electronics SectorThis will includes all the necessary events to start Expansion of technologyand grow the local industry to be the player at higher Once the foundation for compound semiconductor hasvalue-added part in the value chain. For instance, an been established in the short term, Malaysia will beepitaxial process industry will be developed locally for ready to expand the industry to include new state-of-the LED device fabrication. This will mark a shift of the the-art technology in every segments of the industryvalue chain in LED manufacturing from packaging to the with the emphasis to improve the product and bringingepitaxial process to creating the epi-wafer. Figure 3.3 down the cost. This includes wafer manufacturing, LEDshows events for the first term feature.Figure 3.3 Establishing Malaysia’s compound semiconductor industry 41

MEGA SCIENCE 2.0 Electrical & Electronics Sectormanufacturing (Epitaxial growth, device fabrication and sufficiently established to drive the future of compoundLED packaging), and OLED manufacturing. Figure 3.4 semiconductor industry in Malaysia. Meanwhile, severalillustrates the medium-term events. new fundamental technologies will be introduced atNext generation technology the beginning of the long-term plan and targeted to beThe final term (long-term plan) marks the beginning of completed and transform into mass production by thefuture generation technology. In turn, certain technologies end of the term. For instance, phosphorescent organicwhich also have been introduced in the medium term LED (PhoLED) will be introduced and developed,as well as the final term. These technologies will be bringing a brighter future for the LED business. Figure 3.5 shows events for the long-term plan.Figure 3.4 Expansion of new technology for compound semiconductor 42

MEGA SCIENCE 2.0 Electrical & Electronics Sector3.3 GOALS, MILESTONES AND ACTION PLAN The backcasting flow is carefully planned so that to beThe features and events proposed in the previous the market leader in 2050, the final term plan shouldsection are part of the backcasting process. Since the shape the industry towards next generation technology.target in 2050 is for Malaysia to be global leader in To achieve the final term plan, the medium-term shouldsmartphone business through low price and State-of- establish the expansion of new technology and finallythe-art technology, this scenario has been back-casted in order for the medium-term to expand the technology,to three different terms in order to realise the vision. the necessary foundation must be established in the first term.Figure 3.5 Next generation compound semiconductor industry 43

MEGA SCIENCE 2.0 Electrical & Electronics Sector From all the proposed features and events throughout Malaysia should have its own crystal growth processthe terms, the final step in building the roadmap is to not only for compound semiconductor substrate butcreate action plans and milestones to achieve these also for silicon substrate. This involves the process ofgoals. This section will present the necessary action converting raw materials into crystal ingots and finallyplans for three different terms. For all terms, emphasis a sliced wafer. Figure 3.6 shows an overview of thewill be given to achieving low-cost targets while pursuing LED manufacturing industry that covers the total valueadvancement in the technology. chain. In order for Malaysia to establish this industry,3.3.1 SHORT-TERM ACTION PLAN (2013-2020) participation from every segment except manufacturingThe goal for the short term is to establish the compound equipment is required.semiconductor industry in Malaysia. The main ingredient Emphasis will be given to activities to reduce the costin achieving this first term is for Malaysia, especially of the product. The cost reduction of the final productlocal companies, to venture into all other value chain involves an understanding of the source of costs at eachin the industry. In substrate manufacturing industry, key stage in the manufacturing process, and requires careful attention to the design of the product and of the manufacturing process. Figure 3.6 LED manufacturing industrySource: USA Department of Energy, SSL R&D-manufacturing roadmap 44

MEGA SCIENCE 2.0 Electrical & Electronics Sector The typical cost breakdown for an LED package is Figure 3.8 Cost reduction projection of LED manufacturingshown in Figure 3.7. The data represents high vol. Source: USA Department of Energy, SSL R&D-manufacturing manufacturing of 1 mm2 dye on 100 mm diameter roadmapsapphire substrates and packaging of the dye toproduce high power warm white phosphor-convertedLED lighting sources. The Figure indicates that thata significant proportion of the cost contributes by thedye-level packaging stage. Nevertheless, this result isnot surprising as the final product is a packaged dye,and there are many thousands of such dyes on eachwafer (around 5,000 1 mm2 dye on a 100 mm diametersubstrate). Therefore, costs associated with dye-level activitieswill tend to dominate and manufacturers will need toaddress dye-level packaging processes or performmore of the packaging activities at a wafer level inorder to realise the required cost reductions. Figure3.8 demonstrates how the LED package cost elementspredicted to change over time, falling to about 17% of2013 values by 2020. 3.3.1.1 RESEARCH AND DEVELOPMENT In the short term, Malaysia will establish a local compound industry through substrate manufacturing, LED dye manufacturing (epitaxial growth, LED device fabrication, LED packaging), luminaire manufacturing, and MMIC design and fabrication. Table 3.1 shows all the activities for the first term plan, with regards to R&D progress. Figure 3.7 Cost breakdown for an LED manufacturingSource: USA Department of Energy, SSL R&D-manufacturing roadmap 45

MEGA SCIENCE 2.0 Electrical & Electronics Sector Table 3.1 First term roadmap action plan for R&D Goals Action Plans TargetSubstrate manufacturing • Establishing and improving substrate for 1. Production of sapphireEpitaxial growth heteroepitaxial growth (sapphire and silicon) and GaN substrate • Establishing and improving substrate for 2. Bigger substrate homoepitaxial growth (GaN for GaN-on-GaN (>150mm for sapphire LED) and 200mm for silicon) • Establishing Silicon substrate manufacturing • Establishing and improving epitaxial growth process for heterojunction o Wafer uniformity (standard deviation of 0.50nm wavelength for each wafer) o Wafer-to-wafer reproducibility 0.50nm (maximum spread of mean wavelength) o Cost of ownership (COO) 50% reductionDevice fabrication (LED and MMIC) • Establishing and improving device fabrication New GaAs and GaN process process technology o Introduce new compound i. Introduce larger diameter wafers semiconductor process technology o Innovation to reduce cost and ii. Introduce multiple dye improve performances architecture o Deposition of phosphor layer prior Process reduction at chip to wafer dicing level (30% reduction) 46

MEGA SCIENCE 2.0 Electrical & Electronics Sector3.3.1.2 INSTITUTIONAL FRAMEWORK AND POLICIESDevelopment in the compound semiconductor value chain will not be complete without institutions to support theindustry especially for R&D activities. On top of that, some policies must be devised and existing policies may needto be modified in order to enable a smooth transformation in growing the industry. Table 3.2 presents the goals, actionplans, and targets in the first term for institutional framework and policies. Table 3.2 First term roadmap action plans for institutional framework and policies Goals Action Plans TargetResearch and Development • To create R&D centre at national university for At least one R&D centre forcentre every field. the following fields:-Skill development centre Modules for compoundLED adoption as lighting o Substrate manufacturing technology semiconductor industrystandard 40% LED utilisationmega challenge projects o Compound semiconductor fabrication At least one megato solve local issues and (LED, MMIC) challenge initiated by theproblems government as a start-up o New material science project. • To develop curriculum and training for skill 70% local content requirement for all proposed development centre for human capital solutions/products • Governmental policy to insist on utilising LED as the chosen lighting for street and government offices • Government to initiate mega challenge to be taken by local companies/groups to solve country issues/problems. Examples: o Dengue virus rapid detection device o Digital set-top box in preparation for all-digital broadcast o Automatic/robotic system to be used in the palm sector3.3.1.3 INFRASTRUCTURE DEVELOPMENTMoving up the value chain in compound semiconductor industry involves huge capital investment in infrastructure.The front-end part of the industry requires several state-of-the-art facilities including substrate manufacturing plant,device fabrication facility and R&D facility. All these facilities are crucial for participation in front-end activities. Table3.3 presents the infrastructure development needs of the industry. 47

MEGA SCIENCE 2.0 Electrical & Electronics SectorTable 3.3 First term roadmap action plan for infrastructure development Goals Action plans TargetSubstrate manufacturing facility • To develop substrate manufacturing facility for: At least one substrate manufacturing facilityEpitaxial growth facility o Sapphire substrate completed in 2016 o Silicon substrateCompound semiconductor foundry o Gallium Nitride substrate One epitaxial facility byImproving existing LED packaging • To create epitaxial growing facility for the year 2016and luminaire manufacturing plant heteroepitaxial (Sapphire and Silicon) andGLC for sales and distribution homoepitaxial (GaN) MIMOS as one of thecompany as the market for compound • To convert existing CMOS foundry/clean room LED fabrication facilitiessemiconductor industry to include LED fabrication capability. One of the in Malaysia MIMOS facilities can be modified to add LED One compound manufacturing. semiconductor foundry • To create compound semiconductor fabrication by 2016 facility that capable to fabricate LED and MMIC. 30% cost reduction in LED packaging and • Introducing new innovation for existing facility luminaire manufacturing targeting at cost reduction of LED manufacturing GLC based local brand company for • Setting up a company for sales and distribution smartphone business to complete the eco-system of the compound industry (target product: smartphone)3.3.2 MEDIUM-TERM ACTION PLAN (2020-2030)The medium term action plan is to further develop and expand the compound semiconductor industry in the countryto embrace new technology. Application of new innovative technology is important, especially in producing low costproducts. This is to ensure the sustainability and relevancy of the industry and to maintain its competitiveness. Table3.4 shows some of the action plans for the medium term. 48

MEGA SCIENCE 2.0 Electrical & Electronics SectorTable 3.4 Medium-term action plan for research and development Goals Action Plans TargetNew substrate material for LED and • Introduction of plastic material as the 20% of LED device is basedMMIC on plastic and glass material foundation for LED manufacturing. This is forImprove yield for compound flexible LED displaysemiconductor on SiliconIncrease LED utilisation • Introduction of glass material for the LEDLow cost LED product fabrication. This is a preparation for translucent display. • Improve thermal and crystal matching 70% yield achievement on between compound semiconductor and silicon compound semiconductor- substrate. on-silicon devices. • LED lighting penetration towards all street 100% usage on LED lighting lights, government buildings and household by government agencies and 70% usage by household • Innovation in LED substrate, fabrication and 50% cost reduction on LED packaging lamp and luminaire • New process for GaN substrate manufacturing Cheaper GaN based LED device3.3.3 LONG-TERM ACTION PLAN (2030-2050)The final term action plan is themed as the “Next Generation Technology”, suggesting futuristic product development.This action plans are very crucial to achieve the final scenario. The reason behind this is that being a market leaderfor the smartphone industry is challenging. Consequently, such companies will face fierce competition. Therefore, theonly way to be the market leader is by leading the technological advancement faster than the competitor and at thesame time producing low price product. Table 3.5 presents an action plan for the long term. 49

MEGA SCIENCE 2.0 Electrical & Electronics Sector Table 3.5 Long-term Action Plan Goals Action plans TargetNext generation LED displayutilising quantum dot technology • Introduction and integration of quantum 30% penetration of the new LED dot technology into flexible and displayPhosphorescent Organic LED translucent based LED display.technology (PhoLED)Maximum LED utilisation • Introduction of phosphorescent organic New Phosphorescent based organic LED to improve the efficiency. LED with higher efficiency. • Total LED replacement for street lighting, • 100% usage on LED lighting buildings, commercials and household. • 0% usage on incandescent lamp3.4 CONCLUSION The proposed roadmap has targeted interestingMalaysia’s participation in the compound semiconductor scenario in 2050 (the final term). It is planned that byindustry is an encouraging initiative that requires serious 2050, Malaysia will be the market leader for state-of-effort by various parties. The compound semiconductor the-art smartphone industry and some other consumerindustry will continue to grow due to huge demand from electronics businesses. Based on this scenario, severalthe LED display, smartphones, high power applications, feature, events, action plans and targets have beenand telecommunication industries. Thus, it is vital for developed. The roadmap is divided into three terms,Malaysia to gain significant market share in this industry. namely the first term from 2013 to 2020; medium term from 2020 to 2030; and long-term from 2030 to 2050. Figure 3.9 demonstrates the complete Malaysia roadmap for the compound semiconductor industry.Figure 3.9 Overview of compound semiconductor industry roadmap for Malaysia 50

MEGA SCIENCE 2.0 Electrical & Electronics Sector51

MEGA SCIENCE 2.0 Electrical & Electronics Sector 52

MEGA SCIENCE 2.0 Electrical & Electronics SectorCHAPTER 4ENERGY GENERATION, TRANSMISSION AND DISTRIBUTIONBASELINE STUDY: NATIONAL POLICIES, DESIREDOUTCOMES AND INDICATORS, CURRENT STATUS ONLOCAL APPLICATION AND R&D4.1 DEFINITION OF ELECTRIC POWER SYSTEM load centres; and the distribution system that feeds theAn electric power system is a network of electrical power to nearby homes and industries.components used to supply, transmit, and use electric 4.2 COMPONENTS OF POWER SYSTEMSpower as shown in Figure 4.1. One example of an 4.2.1 SUPPLIESelectric power system is the network that supplies aregion’s homes and industry with power. For sizable All power systems require one or more sources of power.regions, this power system is known as a grid and can be For some power systems, the source of power is externalbroadly divided the following segments: the generators to the system, but for others it is part of the system itself.that supply the power; the transmission system thatcarries the power from the generating centres to theFigure 4.1 Basic structure of electric power system 53

MEGA SCIENCE 2.0 Electrical & Electronics SectorThese internal power sources will be further discussed Reactive power does no measurable work but isin the remainder of this section. Batteries, fuel cells, or transmitted back and forth between the reactive powerphotovoltaic cells typically supply direct current power. source and load every cycle. The generators canAlternating current power is usually supplied by a rotor provide this reactive power, but it is often cheaper tothat spins in a magnetic field in a device known as a provide it through capacitors; hence, capacitors areturbo generator. A wide range of techniques have been often placed near inductive loads to reduce currentused to spin a turbine’s rotor, from steam heated using demand on the power system. Power factor correctionfossil fuel (including coal, gas and oil) to nuclear energy, may be applied at a central substation or adjacent tofalling water (hydroelectric power), and wind (wind large loads. Reactors consume reactive power and arepower). used to regulate voltage on long transmission lines.4.2.2 LOADS Reactors installed in series in a power system also limitPower systems deliver energy to loads that perform a rushes of current flow. Small reactors are almost alwaysfunction. These loads range from household appliances installed in series with capacitors to limit the current rushto industrial machinery. Most loads expect a certain associated with switching in a capacitor. Series reactorsvoltage and (for alternating current devices) a certain can also be used to limit fault currents.frequency and number of phases. The appliances found 4.2.5 POWER ELECTRONICSin the average home, for example, will typically be single- Power electronics are semiconductor-based devicesphase, operating at 50 or 60 Hz with a voltage between that are able to switch quantities of power ranging from110 and 260 Volts (depending on national standards). a few hundred watts to several hundred megawatts. The4.2.3 CONDUCTORS classic function of power electronics is rectification, orConductors carry power from the generators to the load. the conversion of AC-to-DC power, power electronicsIn a grid, conductors may be classified as belonging to are therefore found in almost every digital device thatthe transmission system, which carries large amounts is supplied from an AC source. High-powered powerof power at high voltages (typically more than 50 kV) electronics can also be used to convert AC power tofrom the generating centres to the load centres, or the DC power for long distance transmission, in a systemdistribution system, which feeds smaller amounts of known as HVDC. HVDC is used as it has proven to bepower at lower voltages (typically less than 50 kV) from more economical than similar high voltage AC systemsthe load centres to nearby homes and industrial plants. for extreme distancesConductors in exterior power systems may be placed 4.2.6 PROTECTIVE DEVICESoverhead or underground. Overhead conductors are Power systems contain protective devices to preventusually air insulated and supported on porcelain, glass injury or damage during failures. The quintessentialor polymer insulators. Cables used for underground protective device is the fuse. When the current, throughtransmission or building wiring are insulated with cross- a fuse exceeds a certain threshold, the fuse elementlinked polyethylene or other flexible insulation. melts. This produces an arc across the resulting gap4.2.4 CAPACITORS AND REACTORS that is then extinguished; interrupting the circuit. InThe majority of the load in a typical AC power system higher-powered applications, the protective relays thatis inductive; that is, the current lags behind the voltage. detect a fault and initiate a trip are separate from the circuit breaker. 54

MEGA SCIENCE 2.0 Electrical & Electronics Sector4.3 HISTORY OF MALAYSIAN POWER SYSTEM SESB has become one of the subsidiaries of TNB.4.3.1 BACKGROUND OF MALAYSIAN SCENARIO In all three regions, there are also Independent Power Producers (IPPs) supplying some portion of theMalaysia is a country in the Southeast Asia comprised electricity supply to the utility companies to transmit toof Peninsular Malaysia and East Malaysia, separated consumers. Several institutions, including the Economicby the South China Sea. Malaysia’s framework in Planning Unit (EPU), Ministry of Energy, Greenenergy development started when petroleum reserves Technology and Water, the Energy Commission (EC),were discovered in the early 1970s. In 17 August, and Malaysia Energy Centre (Pusat Tenaga Malaysia,1974, Petroliam Nasional Berhad (PETRONAS) was (PTM)) govern the sector. PTM is an importantincorporated under the Companies Act 1964. institution that coordinates various activities, specifically PETRONAS is wholly owned by the Government planning and technological research, development, andof Malaysia and is vested with entire ownership and demonstration in the energy sector (EIA 2013). From thecontrol of petroleum resources in Malaysia through the Economic Transformation Programme (ETP), the EPP14Petroleum Development Act 1974. Other successive Building Transmission and Distribution Companies arepolicies include the National Petroleum Policy 1975, as follows:National Energy Policy 1979, National Depletion Policy a) Transmission and distribution equipment refers1980, Four Fuel Diversification Policy 1981, Fifth FuelPolicy 2000, National Biofuel Policy 2006, and most to equipment used in generation, transmissionrecently, the National Green Technology Policy 2009. and distribution of electricity: switching apparatus, The most important step towards sustainable distribution boards, control panels, transformers,development was the introduction of the Fifth Fuel Policy cables as well as conductors;in 2000.Biomass, biogas, municipal waste, solar and a) Highly values-added and substantial manufacturingmini-hydro were recognised as potential RE sources in of high voltage transmission and distributionelectricity generation. From that time forward, other new equipment; andpolicies have been directed towards utilisation of RE a) Build a cluster of transmission and distributionand promotion of energy efficiency in order to reduce manufacturers in Malaysia: MNCs such as ABB,over-dependence on fossil fuels while also achieving Siemens and Areva, as well as local companies suchsustainable national development. as Tenaga Switchgear, Tenaga Cable Industries and The electricity sector in Malaysia has flourished greatly Malaysian Transformer Manufacturing.in the last ten years. The rapid growth has been hand Hence, a key enabler will be for SIRIM to set up ain hand with the nation’s economic growth, especially high-power testing lab to test high-value, high-voltagewithin the industrial and manufacturing sectors. The equipment. Transmission and distribution products mustelectricity supply industry is vertically integrated with a be certified by an accredited short-circuit testing liaisongenerally monopolistic nature, in which a utility company lab to be accepted by the international market. Mosthandles all the generation, transmission and distribution such labs are in Europe, such as KEMA in Holland andof electricity in region. The main utility companies are CESI in Italy. Due to that, the lab should be operatedTenaga Nasional Berhad (TNB), Sarawak Electricity as a joint venture between SIRIM and an internationallySupply Company (SESCO), and Sabah Electricity accredited lab to be recognised as an accreditedLimited (SESB), each covering the region of Peninsular member (Economic Transformation Programme: AMalaysia, Sarawak, and Sabah, respectively. Roadmap for Malaysia 2010). 55

MEGA SCIENCE 2.0 Electrical & Electronics Sector4.3.2 ENERGY IN MALAYSIA Lunas, Padang Serai, Telok Anson, Langkap, Tampin and Kuala Pilah (TNB 2013).The beginnings of the National Grid gradually materialised On 4 May, 1988, Prime Minister Mahathir Mohamadin 1964 when the Bangsar Power Station was connected announced the government’s decision on a policy ofto the Connaught Bridge Power Station, with the line privatisation. Two items of legislation were passed tosubsequently extended to Malacca. By 1965, a plan was replace the existing Electricity Act and to provide for theset to connect the electricity generating plants spread establishment of a new corporation. Tenaga Nasionalout all over the country. Plants identified for linking were Berhad (TNB) was formed in 1990 under the Electricitylocated at Paka in Terengganu, Temengor, Kenering, Supply Successor Company Act 1990, to succeed theBersia and Batang Padang in Perak, Connaught Bridge, National Electricity Board (NEB) of the states of Malaya.Kapar and Serdang in Selangor, Cameron Highlands 4.3.3 DISTRIBUTION SYSTEMin Pahang, Perai in Penang, Port Dickson in NegeriSembilan, Pergau in Kelantan, Pasir Gudang in Johor Distribution lines of 33 kV, 22 kV, 11 kV, 6.6 kV andand in Malacca. 400/230 Volt in the Malaysian distribution network The central area network with Connaught Bridge connects to the National Grid via transmission substationsPower Station in Klang was the precursor of the energy where voltages are stepped down by transformers.grid. It also tapped into the Cameron Highlands Hydro 4.4 ELECTRIC POWER COMPANIES OF MALAYSIAscheme from the Sultan Yussuf Power Station andextended into a western network. Late in the 1980s, the The main power companies in Malaysia are TNB, SESBloop was finally complete with the placement of Kota and SEB, as shown in Figure 4.2.Bahru within the grid. The Central Electricity Board (CEB) was establishedand began operations on 1 September, 1949. TheBoard was to become heir to three major projectsconsidered by the Electricity Department following its re-establishment in April 1946, which were the ConnaughtBridge Power Station,Cameron Highlands HydroelectricProject & the development of a National Grid. CEBeventually became the owner of 34 power stations witha generation capacity of 39.88 MW, including a steampower station in Bangsar with a capacity of 26.5 MW,a hydroelectric power station at Ulu Langat with acapacity of 2.28 MW, as well as various diesel poweredgenerators with a total capacity of 11.1 MW. On 22 June 1965, the CEB of the Federation of Malayawas renamed the National Electricity Board (NEB) of theStates of Malaya. By the 80s, the Board was supplyingthe whole Peninsular with electricity, strategicallyreplacing the Perak River Hydro Electric Powercompany (PRHEP) and its subsidiary Kinta ElectricalDistribution Co. Ltd (KED) in 1982, Penang Municipalityin 1976. Meanwhile areas supplied by Huttenbach Ltd in1964, which included Alor Setar, Sungai Petani, Kulim, 56

Peninsular Malaysia MEGA SCIENCE 2.0 Electrical & Electronics Sector U Tenaga Nasional Berhad (TNB) Perlis Kelantan Terengganu Labuan Sabah Kedah Sarawak SabahPulau Sarawak Sabah ElectricityPinang Sarawak Energy Berhad Sendirian Berhad )SESB) (80% owned Perak (SEB) by TNB) Pahang Selangor /Wilayah Persekutuan Negeri Sembilan Melaka Johor TNB (P. Malaysia) SEB (Sarawak) SESB (Sabah)• Operates in Peninsular Malaysia • Operates in Sarawak • Operates in Sabah• Total generation capacity is • Total generation capacity is • Total generation capacity is 21,051 MW (2010) 21,051 MW (2010) 866.4 S MW (2010)• Customer is 7,593,684 • Customer is 499,618 • Customer is 413,983 • Max demand: • Max demand: • Max demand: 15,072MW 1036 MW 760 MW• Gen mix (2010): • Gen mix (2009): • Gen mix (2009): - 54% gas - 53% gas - 57% gas - 40% coal - 34% coal - 31% coal - 5.2% hydro - 8% hydro - 9% hydro - 0.2% distillate - 5% diesel - 3% biomassNote: Electricity supply industry in Malaysia is a fully regulated. TNB, SEB and SESB are vertically integrated and operate along with independent power producers (IPPS) Figure 4.2 General profiles of power utilities in Malaysia 57

MEGA SCIENCE 2.0 Electrical & Electronics Sector4.4.1. TENAGA NASIONAL BERHAD (TNB) The Central Electricity Board (CEB) was establishedTenaga Nasional Berhad (TNB) is the largest Electric and came into operation on 1 September, 1949. Theutility company in Malaysia, and also the largest power Board become heir to three major projects considered bycompany in Southeast Asia, with RM69.8 billion worth of the Electricity Department following its re-establishmentassets. It serves over seven million customersthroughout in April 1946, namely the Connaught Bridge PowerPeninsular Malaysia and also the eastern State of Sabah Station of the Cameron Highlands Hydroelectric Projectthrough Sabah Electricity Sdn Bhd, TNB’s core activities and the development of a National Grid. CEB eventuallyare the generation, transmission and distribution of became the owner of 34 power stations with a generationelectricity. Other activities include repairing, testing capacity of 39.88 MW, including a steam power stationand maintaining power plants, providing engineering, in Bangsar with a capacity of 26.5 MW, a hydroelectricprocurement and construction services for power plants power station in Ulu Langat with a capacity of 2.28 MW,related products, assembling and manufacturing high as well as various diesel powered generators with a totalvoltage switchgears, coal mining and trading. Operations capacity of 11.1 MW.are carried out in Malaysia, Mauritius,Pakistan, Indiaand Indonesia.Figure 4.3 Overview of TNB System: Trend of demand growth (2007 ~ 2010) 58

MEGA SCIENCE 2.0 Electrical & Electronics Sector (a) 2000 (b) 2005 (c) 2010Figure 4.4 Final electricity consumption (ktoe) 59

MEGA SCIENCE 2.0 Electrical & Electronics Sector Figure 4.5 Overview Of TNB System:Electricity consumption by customer type (1990~2030)Figure 4.6 Overview of TNB System: Trend of power generation mix (1976~2008) On 22 June 1965, the CEB of the Federation of Kinta Electrical Distribution Co. Ltd. (KED) in 1982. ItMalaya was renamed as the National Electricity Board also took over the supply inthe Penang Municipality(NEB) of the States of Malaya. By the 80s, the Board in 1976, including areas supplied by Huttenbach Ltd inwas supplying electricity to the whole of peninsular; 1964, namely Alor Setar, Sungai Petani, Kulim, Lunas,while strategically replacing the Perak River Hydro Padang Serai, Telok Anson, Langkap, Tampin and KualaElectric Power company (PRHEP) and its subsidiary Pilah. 60