31 July 2016 to 5 August 2016
Gaylord Hotel and Conference Center
US/Eastern timezone

Progress in Coherent Pulse Stacking: A Pathway Toward Compact kHz Repetition Rate LPA Drivers

2 Aug 2016, 13:30
Annapolis 4 (Gaylord Hotel)

Annapolis 4

Gaylord Hotel

Oral Working Group 8 WG8


John Ruppe (University of Michigan)


Next generation LPA drivers producing multiple Joules of ultrashort pulse energy at multi-kHz repetition rates might be based on coherently combined fiber laser arrays. We are developing a temporal multiplexing technique that potentially can reduce the required array size by approximately two orders of magnitude. We demonstrated coherent pulse stacking that enables array-size reduction by an order of magnitude, and are working to extend this technique by at least another order of magnitude.


Next generation laser plasma accelerator drivers will need to produce multiple Joules of ultrashort pulse energy at multi-kHz repetition rates, and with very high wall plug efficiency. Coherently combined fiber laser arrays is a promising path, since fiber lasers can reach very high wall-plug efficiencies. However, limited energy per fiber laser channel could lead to very large array sizes. To overcome this we have proposed an architecture utilizing spatial and temporal multiplexing of fiber lasers, with temporal multiplexing being pivotal for reducing the array size. The current limitation to obtaining large pulse energies from fiber amplifiers is the nonlinearity from the fibers, which limits fiber-based chirped-pulse amplifier energy to approximately two orders of magnitude below its stored energy. In order to extract fully this stored energy, we have proposed the temporal multiplexing technique called coherent pulse stacking. In this scheme, a phase and amplitude modulated burst of pulses is amplified, then stacked into a single pulse using a compact sequence of Gires-Tournois interferometers (GTI), before being compressed using a traditional compressor (e.g. 1ns Treacy compressor). In this way with a sufficiently long pulse burst, 10’s mJ of energy can be extracted from each fiber, thus reducing the number of fibers needed for spatial multiplexing by more than 2 orders of magnitude. So far, we have experimentally demonstrated this technique using a sequence of 4 GTIs to stack a burst of 9 equal amplitude pulses into a single pulse, thus extending CPA by an order of magnitude, and consequently enabling the corresponding reduction in the array size. Further scaling with more interferometers is under way to stack larger numbers of pulses and extend CPA by at least another order of magnitude.

Are you a student? Yes

Primary author

John Ruppe (University of Michigan)


Almantas Galvanauskas (University of Michigan) Hanzhang Pei (University of Michigan) John Nees (University of Michigan) Morteza Sheikhsofla (University of Michigan) Russell Wilcox (Lawrence Berkeley National Laboratory) Siyun Chen (University of Michigan) Wim Leemans (Lawrence Berkeley National Laboratory)

Presentation Materials

Your browser is out of date!

Update your browser to view this website correctly. Update my browser now