Open Access Short Research Article

Predicting Silicon Die Breaking Force in Semiconductor Package Assembly through Mechanical Simulation

Jefferson Talledo

Journal of Engineering Research and Reports, Page 17-25
DOI: 10.9734/jerr/2021/v20i517307

Die crack is a common problem in the semiconductor industry and being able to predict the breaking force at a given loading condition could help prevent such crack problem. This paper presents the use of mechanical simulation in predicting the force at which the silicon die breaks in semiconductor package assembly process. A computer simulation with finite element analysis (FEA) technique was used. The applied force or displacement in a die bending simulation with 3 mm, 4 mm and 15 mm support span was varied until the resulting maximum principal stress of the die becomes equal to its fracture strength. Results revealed that the breaking force for the 70 µm die with 6 mm width is around 5 N for the 3 mm support span and only around 1 N for the 15 mm support span. With the good agreement between modeling and actual results, the study showed that mechanical simulation is an effective approach in predicting die breaking force and can be used to simulate different mechanical loads in the package assembly where possible die crack could happen and be avoided. This is a fast and cost-effective way of assessing risk of die crack and obtaining package assembly process parameters and specifications that are safe to the silicon die.

Open Access Short Research Article

Study of the Impact of Curing Condition on Flexural Strength of a Very Thin Semiconductor Package

Jefferson Talledo

Journal of Engineering Research and Reports, Page 46-51
DOI: 10.9734/jerr/2021/v20i517311

Thinner semiconductor package is becoming popular especially in consumer electronics applications. As package becomes thinner, it is more vulnerable to package crack when subjected to external load. It is important to ensure that the package is strong enough to resist package cracking. This paper presents the study of package flexural strength under different epoxy mold compound curing condition. A 3-point bend test was done to characterize the breaking strength of the package that was subjected to post-mold curing. It was then compared to the strength of the package not subjected to post-mold curing (PMC). Results of the bend testing showed that the package flexural strength is much lower when the package is not subjected to post-mold curing. This study demonstrates that the measurement of flexural strength can be used to determine if the package has undergone post-mold curing or not. Performing the right post-mold curing of the thin molded package is required to ensure higher flexural strength.

Open Access Short Research Article

Modeling Study on the Impact of Mold Thickness on Strip Warpage of a Molded Leadframe Package

Jefferson Talledo

Journal of Engineering Research and Reports, Page 52-58
DOI: 10.9734/jerr/2021/v20i517312

Strip warpage is a common problem in molded leadframe packages. When warpage becomes excessive, the strip could not be processed as it would result in the strip being stuck or damaged during loading to the handling machine loader. This study focuses on the impact of mold cap thickness on strip warpage of a molded Quad Flat No Lead (QFN) package to provide guidance in reducing the strip warpage to an acceptable level. Different mold thickness values were considered in the modeling using finite element analysis (FEA). Results showed that there is an optimum mold thickness, which is around 1.0 mm for 0.20 mm leadframe thickness and 0.65 mm for 0.125 mm leadframe thickness, that gives the lowest warpage in the case considered. The optimum mold cap thickness is lower for thinner leadframe. At mold thickness lower than the optimum value, warpage is in “frowning” mode and increases as the package gets thinner and this agrees with actual observations. This study shows that mold cap thickness has significant impact on molded strip warpage and could be assessed using FEA. Therefore, controlling the mold cap thickness could now be considered an option to reduce strip warpage in molded semiconductor packages as supported by modeling.

Open Access Original Research Article

Indirect Material Cost Reduction by Eliminating Manual Derailing Process through Process Simplification

Christopher L. Dela Cruz, Patricio Cabading Jr., Melanie Paglicawan

Journal of Engineering Research and Reports, Page 1-12
DOI: 10.9734/jerr/2021/v20i517305

This paper will discuss how to reduce (IDM) Indirect Material consumption for derailing process by eliminating the manual derailing method as part of cost saving project. It involves the removal of cutter blade used for derail.Cost improvement was one of organizational goals of the company for 2019. This drive can be supported by analyzing Indirect Material spending and process simplification at assembly plant. At assembly End-Of-Line processes, derail cutter blade was one of the top Indirect Material spending at singulation and there is an opportunity to reduce if not eliminated the cost consumption of derail cutter blades through process simplification. Increasing volume in Quad-Flat-No lead (QFN) packages and new banner products being develop by New Product Integration NPI, Q1’20 means increase in IDM consumption per process. The challenge is to drive a process simplification that will reduce IDM to save cost by start of Q2’20. DMAIC methodology was used to improve the process of derailing process.

Open Access Original Research Article

Eliminating Frequent Machine Assists through Process Plate Enhancement on Wirebonding Process

Jonathan Pulido, Edwin Graycochea Jr., Frederick Ray Gomez

Journal of Engineering Research and Reports, Page 13-16
DOI: 10.9734/jerr/2021/v20i517306

New devices and technologies in the semiconductor industry are getting more challenging to process because of inherent issues especially on quad-flat no-leads (QFN) packaging assembly. This paper is focused on the improvement done for QFN leadframe device to address the major machine assist during the lot processing at wirebond process. Illumination and visual of the leadframe and the sand blasting process plate on the machine are difficult to separately distinguish due to similar color shade of the materials, thus frequent machine assists ensued. To reduce the frequent machine assist occurrence, an improvement is done through enhancing the process plate by using a black chrome to totally separate the illumination of leadframe and the process plate. Ultimately, the machine assist during wirebonding process is improved (the longer the better) from 16 minutes to 6 hours continuous and uninterrupted running.

Open Access Original Research Article

Introduction of Laser Pi-Grooving as Breakthrough Solution to Enhance die Strength of 40 nm ulow-k CMOS Silicon Technology during Wafer Saw Process

Aiza Marie E. Agudon, Hynlie B. Inguin, Bryan Christian S. Bacquian

Journal of Engineering Research and Reports, Page 26-36
DOI: 10.9734/jerr/2021/v20i517308

Nowadays, semiconductors and electronics are becoming part of our everyday activities. As the Integrated circuits become more useful to people, it also requires more function, which contain more complex and compact components. Aligned to this package requirement, the more challenging it become to package development as Silicon technology becomes more critical and complex from bare silicon to conventional MOS technology to Ultra Low-K, which requires a different strategy.  The new process development in the Semiconductor industry is a necessity to cope up with these new technologies. Low-k devices always pose a big challenge in achieving good dicing quality. This is because of the weak mechanical properties of the low-k dielectric material used.  Mechanical Sawing is the most popular cutting method for silicon, but with Ultra low-K technology, using mechanical sawing will lead to various sawing defects such as chippings and delamination [1,2]. These leads to the introduction of Laser Grooving to get rid of these dilemmas. Laser grooving uses heat to eradicate metals on this very thin metal wafer dicing saw streets in preparation for wafer saw process to prevent topside chippings and delamination/metal peel off [3]. These defects are not acceptable especially since the product application is a chip card. Since chip cards must be flexible and durable, they require higher die and package strength to serve its purpose. To achieve such package requirement, different method was evaluated such as standard mechanical dicing, standard Laser Grooving and the PI laser groove.  

The paper will discuss how we were able to achieve the quality requirement for Ultra Low-K and at the same time eliminating top reject contributor during startup of this device.

Open Access Original Research Article

Improvement on QFN Leadframe Design of Extended Leads to Support the Mitigation of Mold Flash Occurrence

Alyssa Grace Gablan, Frederick Ray Gomez, Anthony Moreno

Journal of Engineering Research and Reports, Page 37-40
DOI: 10.9734/jerr/2021/v20i517309

Parameter optimization is not only the key to find the most favorable and best solution as variable chances of failure modes may happen at extreme case conditions at unexpected period. Packaging design robustness is much resilient to establish a satisfactorily good quality product and sustain a long-term goal of a remarkable process capability. This paper presents leadframe            design solution of quad-flat no leads (QFN) to address mold flash defect caused by leadframe bouncing during wirebonding. An inverted pyramid configuration was conceptualized to              provide better stability than the standard configuration during wirebonding process and other concerned assembly processes due to the shift of the center of gravity moving closer to the full metal part.

Open Access Original Research Article

Die Placement Performance Evaluation on Different Machine Platforms

Rennier Rodriguez, Edwin Graycochea Jr., Frederick Ray Gomez

Journal of Engineering Research and Reports, Page 41-45
DOI: 10.9734/jerr/2021/v20i517310

The paper focuses on the die placement tolerance capability with critical design for land grid array (LGA) device evaluated on two different die attach machine platforms. The evaluation was narrowed down into two main die attach machines with the objective of attaining the best performance in terms of die placement tolerance capability. Die placement tolerance were quantified and only die attach Machine B was able to satisfy the specification of less than 30 microns. The study used a bar graph comparison in terms of die placement tolerance on the two machines and presented the effect of machine selection on die placement tolerance capability. For future works, the selected die attach machine could be used for devices with critical requirement.

Open Access Original Research Article

Die Attach Process Robustness through Epoxy Pattern Optimization for DFN Device

Jerome J. Dinglasan, Frederick Ray I. Gomez

Journal of Engineering Research and Reports, Page 59-63
DOI: 10.9734/jerr/2021/v20i517313

As the modern world evolves through innovation and technology, manufacturing companies of semiconductor drive their people to dedicate themselves on continuous improvement and technical advancement. Products they produce becomes smaller and thinner, but applications are limitless and innovative. In manufacturing of quad flat no lead packages, challenges were encountered during the die attach process especially in small die size. Die attach material overflow on pad is one of these, and certain parameters and techniques are explored to have a quality and robust process. Defining appropriate epoxy pattern with respect to the die size is a big factor to make sure no epoxy material will flow outside the die pad perimeter that may cause unit rejection. This paper will discuss the related issues of the said package by performing experiments and applying certain techniques to address the problem.

Open Access Original Research Article

Comparative Analysis of Some Existing Models for Estimating the Time of Concentration for Watersheds in Anambra State, Nigeria

J. C. Agunwamba, N. C. Mmonwuba

Journal of Engineering Research and Reports, Page 64-75
DOI: 10.9734/jerr/2021/v20i517314

This study considered the estimation of the time of concentration (TC) using 30 different watersheds in Anambra State, Nigeria. The study assessed the performance of some existing models for the estimation of time of concentration in the study area. Data for this research were collected from watersheds located at Awka-South Local Government in Anambra State. A measured time of concentration values was also obtained by using a tracer at the watershed divide and the time it took the tracer to get the outlet of the watershed was recorded. This was carried out on all the 30 watersheds at the same time. The length, slope and area of the 30 watersheds were also measured. Thereafter, the time of concentration was estimated using the 30 existing models. The extent of linear association between the observed and estimated time of concentration for the different models was determined. The outcome of the study revealed that the Ventura model (Tc12) developed in Italy recorded the highest correlation coefficient with a measure of 0.681, followed by DNOS model (Tc14) with a coefficient measure of 0.661 while the least performing model was Picking model (Tc13) with -0.423 correlation measure. There was also an obvious difference in the values of the time of concentration calculated using the different models. Therefore there is always a need to verify any model to be used for estimating the time of concentration in other to have a more robust design. 

Open Access Original Research Article

Flood Flow Probability Distribution Model Selection on Niger/Benue River Basins in Nigeria

Itolima Ologhadien

Journal of Engineering Research and Reports, Page 76-94
DOI: 10.9734/jerr/2021/v20i517315

Flood frequency analysis is a crucial component of flood risk management which seeks to establish a quantile relationship between peak discharges and their exceedance (or non-exceedance) probabilities, for planning, design and management of infrastructure in river basins. This paper evaluates the performance of five probability distribution models using the method of moments for parameter estimation, with five GoF-tests and Q-Q plots for selection of best –fit- distribution. The probability distributions models employed are; Gumbel (EV1), 2-parameter lognormal (LN2), log Pearson type III (LP3), Pearson type III(PR3), and Generalised Extreme Value( GEV). The five statistical goodness – of – fit tests, namely; modified index of agreement (Dmod), relative root mean square error (RRMSE), Nash – Sutcliffe efficiency (NSE), Percent bias (PBIAS), ratio of RMSE and standard deviation of the measurement (RSR) were used to identify the most suitable distribution models. The study was conducted using annual maximum series of nine gauge stations in both Benue and Niger River Basins in Nigeria. The study reveals that GEV was the best – fit distribution in six gauging stations, LP3 was best – fit distribution in two gauging stations, and PR3 is best- fit distribution in one gauging station. This study has provided a significant contribution to knowledge in the choice of distribution models for predicting extreme hydrological events for design of water infrastructure in Nigeria. It is recommended that GEV, PR3 and LP3 should be considered in the development of regional flood frequency using the existing hydrological map of Nigeria.

Open Access Original Research Article

Boolean Subtraction and Division with Application in the Design of Digital Circuits

Okoh Ufuoma

Journal of Engineering Research and Reports, Page 95-117
DOI: 10.9734/jerr/2021/v20i517316

The subject of Boolean subtraction and division dates back for over a century to the work of George Boole. Nevertheless, this subject is unfamiliar to us because it has been banished from Boolean algebra. In fact, some authors claim that there is no such thing as Boolean subtraction and division. The purpose of this work, however, is to present with clarity the subject of logical subtraction and division and its practical application in the design of digital circuits.

Open Access Original Research Article

Small Pilot System for the Manufacture of in Situ Precipitated Calcium Carbonate in the Presence of Pulp Fibers

Klaus Dölle, Bardhyl Bajrami

Journal of Engineering Research and Reports, Page 118-123
DOI: 10.9734/jerr/2021/v20i517317

Paper fillers materials are less expensive than fiber, allowing reduced production costs and improvement of paper properties.

A small pilot scale in-situ laboratory calcium carbonate filler precipitation unit was developed and designed with the objective to provide enough pulp fiber containing in-situ precipitated calcium carbonate for a small 12-inch (304 mm) wide laboratory paper machine. The in-situ precipitation system requires the reactants calcium hydroxide and carbon dioxide and was tested with a eucalyptus pulp fiber pulp suspension.

The final precipitated in-situ filler content achieved was 38.2%, 55.5% and 66.6% based on initial eucalyptus pulp fiber content. The precipitation time from an initial pH of 12.77, 12.76 and 11.98 to an final pH of 7.29, 7.55, and 7.28 for the 3 kg, 6kg, and 9 kg of calcium hydroxide reactant addition was 45 minutes for the 3 kg and 9 kg calcium hydroxide addition and 40 min. for the 6 kg calcium hydroxide addition.

Open Access Original Research Article

Investigation of the Adsorption Potentials of an Organic Adsorbent for Phenol Removal from Aqueous Solution

C.E. Muko-Okoro, I.A. Obiora-Okafo, J.N. Ndive

Journal of Engineering Research and Reports, Page 124-134
DOI: 10.9734/jerr/2021/v20i517318

Phenol is a prevalent pollutant found in many industrial wastewaters, and it is paid singular attention because of its special features including high toxicity, carcinogenic properties, and vital cumulative ability that affects the health of humans and the environment. The current study investigated the removal of phenol from synthetic aqueous solutions using prepared Moringaoleifera seed shell as an adsorbent. The efficiency of phenol removal by Moringaoleifera seed shell was evaluated in a batch system, and different parameters such as initial concentration of phenol (100, 200, 300, 400 and 500 mg/L), contact time (10, 20, 30, 40, 50 and 60 min), and adsorbent dosage (0.2, 0.4, 0.6, 0.8, and 1.0 g) were studied. The results showed that the highest percentage of phenol removal by the ash occurred at 0.8 g dosage, contact time of 40 min, and initial concentration of 500 mg/L giving 87.2% phenol removal. The adsorption process was modeled with Langmuir and Freundlich isotherms and adsorption kinetics (pseudo-first order and pseudo-second order) at controlled temperatures. The results showed that the experimental data fitted the Langmuir (R2 = 0.8338) much better than the Freundlich model (R2 = 0.7314). For the analysis of the adsorption kinetics, the results showed that the experimental data fitted the pseudo-second order kinetics (R2 = 0.999) much better than the pseudo-first order kinetics (R2 = 0.5042). In general, the results of this study revealed that Moringaoleifera seed shell has suitable potential for use in removing phenol from aqueous solution on operation and practical scales due to its availability and organic nature.

Open Access Original Research Article

Biochemical Methane Potential Assessment by Anaerobic Digestion of Locally Available Grasses of Phuentsholing, Bhutan

Yeshi Choden, Samten Zangmo, Saahin Tamang, Thinley Gyeltshen, Karma Phuntshok, Sanjita Limboo, Tenzin Choden Gyeltshen

Journal of Engineering Research and Reports, Page 135-144
DOI: 10.9734/jerr/2021/v20i517320

Renewable energy is not only environmental friendly but also promotes sustainable development. Biogas being one of the abundantly used renewable resource, the enhancement and optimization of the yield of biogas can help in reduction of dependence on imported fuel. Biochemical methane potential (BMP) assessment of grass will determine the production of methane (CH4) from this substrate through the process of anaerobic digestion. After determining the parameters such as pH, Biochemical Oxygen Demand (BOD) and Total solids (TS) of three types of local grasses known as Basil, Bermuda and Napier, that affects the production of biogas, Napier grass resulted with the highest potential to produce CH4 gas. Batch and continuous reactor method under mesophilic condition was adopted. The composition of biogas from continuous reactor was obtained using a biogas analyzer (Biogas 5000 Geotech), from which 30.8% of CH4,

 8% of CO2 and other inert gases were found. Also, methane to carbon dioxide (CH4: CO2) ratio of 3.81: 1 approximately (80% - 20%) was achieved. Moreover, the batch reactor method showed that 1L Napier grass silage would yield 0.81L of biogas. The concentration of CH4 gas from Napier grass in hydraulic retention time as short as 20 days was very significant. This study shows that Napier grass can be used as an alternative sustainable source of energy in the country which can improve resource constraints.