Technologies used to provide cooling for buildings at U.S. Department of Defense (DoD) sites represent a substantial portion of the energy consumption at fixed installations and forward operating bases. Conventional vapor compression and absorption cooling systems involve refrigerants and chemicals that require special handling to prevent toxic exposure or harmful discharge into the environment. With increasing energy costs, the economic and security impacts of cooling loads are an important consideration when planning for retrofit improvements and new installations.

The objective of this project was to demonstrate the technical and economic efficacy of a solar-thermal cooling system using an adsorption chiller at the Parris Island Marine Corps Recruit Depot (PIMCRD) in South Carolina. Quantitative and qualitative performance objectives included demonstrating peak cooling capacity, maximum capacity using solar energy alone, steam and electrical energy reductions and associated emissions footprint reductions, operability and reliability, and economic benefits.

Technology Description

The primary components of the demonstrated technology are an adsorption chiller, an array of evacuated tube solar panels, and balance of plant equipment (e.g., pumps and piping) necessary to support operation of the system as integrated with existing heating, ventilation and air conditioning (HVAC) equipment at the site (e.g., air handlers and controls). The system design also included the ability to utilize excess solar thermal energy (e.g., in winter months) for domestic water heating.

Adsorption chillers are better suited than absorption chillers for solar thermal applications because of their ability to operate over a wider range of hot water supply temperatures (as low as 120°F). Adsorption chillers, like all chillers, extract heat from the environment by way of vaporization of a refrigerant liquid. In an adsorption chiller, the refrigerant (water) is evaporated under vacuum conditions. It is then adsorbed (condensed) onto a solid sorbent; in this case, silica gel. The silica gel is regenerated (desorbed) using hot water supplied by solar energy and/or steam.

Evacuated tube solar panels were employed because of their higher efficiency and higher output temperatures compared to flat panel collectors.

The demonstration was conducted at the 1st Battalion Mess Hall (Building 590) located at PIMCRD. The baseline cooling system was an absorption chiller (90 tons refrigeration [RT]) powered by steam for cooling the dining area. An electric chiller (60 RT) was used to cool the kitchen area.

The conceptual test design was to evaluate the performance of the existing system before modification as a baseline and compare this to the performance of the modified system to determine energy savings. In addition, the performance of major sub-components of the system was to be evaluated. Major sub-components included the adsorption chiller, the solar array loop, and the domestic hot water (DHW) system.

Demonstration Results

The performance objectives, success criteria, and demonstration results are summarized in the table below. None of the demonstration objectives were fully satisfied. Although operation of the adsorption chiller was not demonstrated at full rated capacity, analysis shows that this limitation resulted from balance of plant design and implementation issues, not inherent problems with the adsorption chiller. The solar array performed exactly as expected. Due to construction delays, system design, and site integration issues, the DHW system was not operational prior to the end of the demonstration. Additional details that impacted demonstration results are discussed in the following section on implementation issues.

Performance Objectives and Outcomes

Performance Objective

Success Criteria


Quantitative Performance Objectives

Peak cooling capacity of the System Under Test (SUT)

Peak cooling capacity of SUT must be greater than 80 RT.

Objective not met. Maximum sustained cooling was 52.8 RT.

Max cooling capacity of the SUT when driven by solar energy only

When DHW solar energy demand is zero, peak cooling capacity of SUT must be greater than 60 RT without supplemental steam.

Objective could not be evaluated. The final system design did not permit the system to be operated on solar output alone.

Steam energy reduction

Steam energy reduction will exceed 800 MMBtu/year including cooling and DHW.

Objective not met. Annualized net steam energy reduction over baseline was 703.8 MMBtu/year.

Emission foot print reduction

Reductions exceed 79 metric tons CO2e per year relative to baseline.

Objective not met. Net GHG emissions reduction was 32.1 metric tons CO2e per year.

Equipment availability and reliability

>99% availability >99% reliability

Objective not met. Overall system availability and reliability could not be quantitatively assessed due to ongoing operational issues throughout the demonstration.

Economic performance

Simple payback < 7 years; Positive NPV based on ECAM and BLCC

Objective not met. Site-specific payback will not be achieved within the system lifetime.

Ease of use

The average points above neutral (or above three points)

Objective could not be fully evaluated as the system did not achieve stable, routine operations during the demonstration period. The system is complex and unfamiliar. Training and documentation were incomplete as of the end of the demonstration. Implementation issues complicated operations.

BLCC = Building Lifecycle Cost

CO2e = carbon dioxide equivalent

ECAM = Environmental Cost Analysis

GHG = greenhouse gas

MMBtu = million British thermal units

NPV = net present value

Implementation Issues

Unexpected technical and management issues were encountered during the course of the project that negatively impacted the outcome of the demonstration. Significant issues included:

  • The available roof area was insufficient to support the planned solar thermal capacity.
  • The initial design for chiller operation on solar or steam energy alone was unworkable due to insufficient solar capacity to operate the chiller on solar energy alone as well as piping design issues that prevented operation on steam without utilizing the solar buffer tank.
  • The piping design failed to account for normal water transfer between the hot water and tower loops within the chiller.
  • The solar field piping construction, though built to manufacturer specifications, was inadequate to withstand high temperatures during stagnation events.
  • The initial piping design had inadequate provision for pressure relief and release of entrapped air.
  • The initial control sequence was incomplete, which caused delays in system commissioning.
  • The design failed to make adequate provisions to ensure that chiller supply flows and temperatures would meet the chiller submittal specifications.

The adsorption chiller factory acceptance test conditions matrix did not anticipate the range of possible supply flow and temperatures to the chiller that might be encountered in the field, or span the range of chiller cycle timing that might be employed to optimize performance under field conditions. This made it impossible to quantitatively determine whether the chiller performance in the field was within the expected range, and complicated efforts to optimize chiller performance in the field.

There were also issues with building HVAC systems maintenance that negatively impacted the ability to fully evaluate the performance of the test system. In particular, during much of the 2012 cooling season, maintenance issues with the electric chiller (installed in series with the adsorption chiller) affected adsorption chiller performance such that results were not representative of normal chiller performance.

Based on experience with this demonstration and findings from other researchers, the capital cost of a solar thermal chiller system using evacuated tube collectors is unlikely to be recovered from energy savings alone. A thorough design effort accounting for net parasitic loads, piping friction head, and building HVAC system operating details would be required to achieve payback within the system lifetime (20-30 years).

  • Chiller,

  • Water Heating,

  • Renewable Energy,

  • Solar,