Review A comparative study of charge pumping circuits for flash memory applications O.Y. Wong ⇑ , H. Wong, W.S. Tam, C.W. Kok Department of Electronic Engineering, City University of Hong Kong, Tat Chee Avenue, Kowloon, Hong Kong, China article info Article history: Received 15 September 2011 Accepted 30 September 2011 Available online 1 November 2011 abstract Flash memories are now widely used in many portable electronic devices, in embedded systems and are even as replacement for computer hard disks. In flash memory systems, high-voltages (up to about 10 V) are indispensable for programming operations. In many cases, however, such programming voltages are not directly available from the supply, and are usually generated by embedded voltage converting or charge pumping circuits. These circuits produce the required programming voltage from available exter- nal supplies with voltages in the approximate range of 1–5 V. The power conversion efficiency, the chip size, the voltage regulation, as well as the loading characteristics have been the major concerns for such circuits. The present paper discusses some recently proposed charge pumping circuits for flash memory applications. We focus on the effects of the dynamic gate control, the 4-phase gate-boosting and cross- coupled configuration for enhancing the performance of the charge pump circuits. Several different charge pumps operated under different working conditions are then investigated in detail. Ó 2011 Elsevier Ltd. All rights reserved. Contents 1. Introduction ......................................................................................................... 670 2. Overview on existing charge pump circuits ................................................................................ 672 2.1. Dynamic inverters ............................................................................................... 673 2.2. Gate boosting circuit ............................................................................................. 673 2.3. Cross-coupled structure .......................................................................................... 674 3. Characteristics comparison ............................................................................................. 675 3.1. Simulation details ..................................................................................................... 675 3.2. Simulation results ............................................................................................... 677 3.3. Discussion ..................................................................................................... 677 3.3.1. Output voltage ................................................................................................... 677 3.3.2. Power efficiency .................................................................................................. 682 3.3.3. Area ............................................................................................................. 683 3.3.4. Process requirement and reliability problem ......................................................................... 684 3.3.5. Cascading capability ................................................................................................... 684 4. Negative charge pump and other charge pump structures .................................................................... 686 5. Conclusion .......................................................................................................... 686 Acknowledgment ..................................................................................................... 686 References .......................................................................................................... 686 1. Introduction Flash memory is a dominant non-volatile memory component nowadays. Its attracting features includes small size, high density, high robust, lost cost and simple accessory circuits. It has been widely used in portable electronic devices like cellular phones, digital music player and digital cameras, for many years. As its fabrication process is compatible with that of the mainstream CMOS technology, it has also been used in many embedded sys- tems and micro-controllers for microcode storage. Recently, the use of flash memory as solid-state drive (SSD) to replace the hard disk in personal computers has also become very popular. SSDs were also introduced into high-end systems such as enterprise 0026-2714/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.microrel.2011.09.031 ⇑ Corresponding author. E-mail address: oi_ying1009@yahoo.com.hk (O.Y. Wong). Microelectronics Reliability 52 (2012) 670–687 Microelectronics Reliability journal homepage: www.elsevier.com/locate/microrel