In fires, corrosive chemicals are generated along with the other combustion products, which are present as gases, liquids and solids.1 Corrosive chemicals (such as HCl, HBr, and HF) are commonly emitted as gases and liquids with inorganic atoms in the structure. Non-corrosive combustion gases and liquids are emitted as organic compounds and water, whereas solids are emitted as soot, inorganic metals, and dust. As a result, corrosion caused by smoke generated after a fire is due to the presence of acid gases (e.g., HCl, HBr, and HF) in smoke plus moisture/water in the environment.2,3 

Due to the stringent requirements to control temperature, moisture, and contamination (gaseous and particulate) for data center operation, the smoke environment created after a fire can be highly corrosive for IT equipment in areas encountering high concentrations of smoke. Electronics contaminated by smoke (i.e., soot, particulates, and acid gases) will experience damage either short or long term after the fire by circuit bridging and corrosion-related failures, especially when halogenated plastics are burned.4 

To evaluate smoke corrosivity of representative components (containment panels, air filters, and electrical power cables) in data centers, a tube furnace system connected to a gas bubbler assembly was used to perform combustion testing and collect corrosive acid gases generated by the combustion of these materials. Analytical techniques including ion chromatography (IC), pH, solution conductivity, and linear polarization resistance (LPR) were applied for smoke corrosivity assessment. The performance requirements/thresholds are specific to the experimental setup and parameters for this setup were established. These parameters are based on the comparison to known corrosive or noncorrosive smoke from previous studies.1,2,3,5 

This paper will present the testing techniques developed to evaluate the corrosivity of smoke generated from the combustion of data center components after fire incidents in data centers. The performance requirements/thresholds derived in this paper will advance the corrosion community to mitigate smoke corrosivity issues and reduce equipment losses after fire incidents.

Subject
Test methods, Materials, Corrosion rate, Corrosivity, Tubes, Aqueous solutions, Acidity, Cables, Furnaces, Filters, Corrosion data, Linear polarization resistance, Conductivity
Keywords:
Smoke Corrosivity, Fire Smoke, Combustion Products, Data Center Components
© 2020 Association for Materials Protection and Performance (AMPP). All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means (electronic, mechanical, photocopying, recording, or otherwise) without the prior written permission of AMPP. Positions and opinions advanced in this work are those of the author(s) and not necessarily those of AMPP. Responsibility for the content of the work lies solely with the author(s).
2020
Association for Materials Protection and Performance (AMPP)
You do not currently have access to this content.