What's new in LIGO

David Shoemaker, LIGO-MIT dhs-at-ligo.mit.edu
Since the last MOG, the LIGO Laboratory and more broadly the LIGO Scientific Collaboration (LSC) has been working on technical issues in both data analysis and in instrument science. This note will concentrate on the instruments, complementing the last MOG LIGO report which described the first observation publications.

Initial LIGO

The LIGO interferometers, installed in the observatories in Livingston, Louisiana, and Hanford, Washington, have interleaved observation with commissioning over the past few years. Since completing the S3 run in January of 2004, all the interferometers have been going through a mixture of tuning and the addition of new elements to bring them to the desired sensitivity.

An important step forward has been the commissioning of the Hydraulic External Pre-Isolator, at the Livingston observatory. This system was originally designed as an element of the Advanced LIGO seismic isolation system, but was pressed into early application to reduce the excessively large ground motion in the 0.2 - 10 Hz band at Livingston. It is an active seismic isolation system, using inertial sensors and actuators in all six degrees of freedom to reduce the motion of the structure supporting the original seismic isolation `stack'. It delivers about a factor of ten reduction in motion - enough to permit the Livingston interferometer to lock during the day and through the passage of trains on a nearby track. This should increase the uptime of the detector, and also allows commissioning during the day.

Another new element in the interferometers is a Thermal Compensation System. This is again an Advanced LIGO element brought to bear on initial LIGO. The notion is to deliver heat to the interferometer optics to change their focal length (via $dn/dt$ ). This can be used on `cold' optics to compensate for an initially slightly incorrect radius of curvature, and/or to compensate for excessive focusing in a `hot' optic already distorted by the main sensing laser beam; Gaussian or `doughnut' profiles, or more complex forms for spatially varying absorption, are possible. Using the TCS, significant improvements in the quality of the modulation sidebands have been made along with reductions in feedthrough of oscillator phase noise. We are still installing the TCS system, and still learning how to use it, but it has already and will clearly lead to further reduced shot noise and other sensing noises.

A variety of other details have been worked to reduce the noise contributions. Higher gain control loops for the `auxiliary' servo loops, improved methods to balance the actuators on the suspensions, modifications of the filtering in photo-detector amplifiers, and lower-noise oscillators for the modulation/demodulation systems have all helped. The best sensitivity to date is in the Hanford 4k instrument, and is shown in Figure 1. The sensitivity is within a factor of two of the goal across the entire target sensitivity of the instrument, and we understand well the remaining limits. We are working on propagating the improvements to all interferometers.

Figure 1: Caption: The strain sensitivity of the LIGO Hanford 4km gravitational wave detector, showing its evolution with continued commissioning. The bottom-most measured curve dates from August, 2004. The smooth line at the bottom is the goal for the sensitivity of the instrument, as laid out in the Science Requirements Document (`SRD'). (LIGO G040439-00)



The plan for further data collection and improvements calls for a one-month observation run, S4, to start in early 2005, followed by a push to bring all the instruments to the goal for the initial LIGO instruments. Then the S5 run, currently planned to start late in 2005, is targeted to collect one year of integrated observation with the initial LIGO detector.

Advanced LIGO

The other significant effort in instrument science is to bring Advanced LIGO forward. An important milestone was passed in October, when the National Science board reviewed the Advanced LIGO proposal. They recommended to the director of the NSF that Advanced LIGO be funded as requested. There are certainly significant hurdles yet to be passed before funding is received, but this is a necessary and very important step toward the realization of Advanced LIGO.

A number of the subsystems have made nice advances recently. The pre-stabilized laser, led by the Albert Einstein Institut Hannover, saw their partners Laser Zentrum Hannover achieve the required 200 watts of laser power from the prototype laser power head for Advanced LIGO. The high-power test facility at Livingston came on line, and the Input Optics subsystem led by the University of Florida started tests of modulator and isolator materials at realistic power levels. Efforts in the LIGO Laboratory included optical coating development, which explored parameter space of dopants and found lower mechanical losses, important for the thermal noise. Both the requirements for optical losses and uniformity for the thermal compensation, and successes in making coatings meeting them, were realized. A full mode-cleaner style prototype from the Caltech suspension group was installed at the MIT LASTI facility and characterized, and the quadruple suspension development, led by the UK/Glasgow with lots of Caltech/MIT participation, moved forward. The prototype isolation system at Stanford, with LSU's leadership, was prepared and then operated in vacuum with new control laws. The 40m interferometer configuration testbed at Caltech was able to lock all degrees of freedom, leading the way to tests of the locking of Advanced LIGO and comparison with models.

One significant point to mention is the choice of substrate for Advanced LIGO's test mass optics. We had been looking very carefully since 2001 at two materials: fused silica, which is the traditional material for fine optics, and used in initial LIGO; and sapphire, a very hard, high density, low-mechanical loss material. Careful consideration of both the performance measures (e.g., they exhibit different thermal noise `signatures', and different net noise levels for a given coating thermal noise), and practical questions (e.g., the ability to manufacture and install complete systems on a schedule) were taken into account. Curiously, the cost for either option was the same - so this was not a net criterion. The Lab has adopted the recommendation from this study group to use fused silica as the baseline material, and this allows the quad suspension group to move forward with dimensions and density for the test mass. A closing note on this choice is that, since the internal thermal noise is considered to be quite low for fused silica, any extra improvements in coating thermal noise will lead to similar improvements in the Advanced LIGO sensitivity. A nice challenge!

The coming year will see further full-scale prototyping and testing, development of readout systems for the interferometer testbed, development of the complete Advanced LIGO model in the `e2e' package, and further progress in the other subsystems. And, we hope, good news on the Advanced LIGO funding timescale.

Jorge Pullin 2005-03-10