Reliability Engineering Snapshot TM

Illustrated Case Studies in the Maintenance Reliability Engineering World of Failure Analysis, Predictive Maintenance, and Non Destructive Evaluation




Material Properties - Case Study No. 104: High Temperature Mu Phase Embrittlement of C276


Overview of C-276 Nozzle

Is it erosion or is it high temperature oxidation, or is it something that lurks below the surface? If you have a component in high temperature service that "wears out" in an erratic manner and frequency, don't be too quick to blame it on the operators or the mechanics, the machine shop that made it, or foreign material suppliers. It may be a subtle misapplication of the material for its intended service.

The nozzle shown to the left, had a good service life history. It was replaced periodically every several months or so. The nozzle was in a high temperature application. The region of very high temperature for this nozzle, relative to the rest of itself, was at the small opening at the top. During operation, a combustible mixture streamed from this opening and immediately ignited. Over a period of time under normal service conditions the small opening would enlarge and throw the tolerances out of limits; this in turn affected the product quality which was continuously monitored. The nozzle would then be replaced. This was how life was defined for this component over the last 13 years. Recently however, the nozzle was wearing out "faster than normal." As is often the case "the usual group of suspects were rounded up" for interrogation and accusation.

In this case the blame was place upon the machine shop that made it. The accusation was that they were doing something different to cut costs and it was showing up in the poor performance of the nozzle. Initial inspection of the metal around the opening showed signs of intergrannular attack. You can see it closer up by clicking on the opening in the picture above. Looking at that close up, there was a noticeable difference between one end of the opening and the other end. Although the close up picture does not do justice, there was noticeable intergrannular cracking in the upper left region, as compared to the lower right region which was clean. Since this material had been successfully used for quite some time it was felt that the material was somehow corrupted. Therefore, fuel was added to the fire when it was discovered that there was "copy cat material" out there (i.e. equivalent material made by someone else), which in turn automatically meant that it wasn't as good as the real stuff no matter how unsubstantiated the reasoning. This information was all some people needed to hear.

The nozzle was fabricated by an outside company. Literally speaking, the cross hairs of the rifle scope were sighted right upon the machine shop foreman's forehead. People figured that the foreman had switched to a cheaper copy cat material in order to lower fabrication costs and make more money. Before the trigger was pulled, and a good business relationship was lost, the decision was made to analyze the material. At least that way it was felt the findings would show "just cause" for the grievous action taken. Two nozzles were sent out for analysis. It was believed by all that the analysis would confirm that the material was an inferior copy cat material.



The material was C-276. It is a nickel-chrome-moly alloy material which was first introduced back in 1965. The picture to the left (200X) shows the microstructure of a good sample known to be Hastelloy C-276. Quantitative EDX of the suspect material and the known C-276 reference material were nearly identical.





There was an anomaly though. As can be seen in the picture to the left (1,000X), there were precipitates. What was different about these precipitates? EDX showed that they contained a higher amount of molybdenum and tungsten, on the order of two to three times more, respectively, than the composition of the good reference sample. To add to the mystery, the precipitates contained about the same amount of chromium as the good sample. This cast a shadow of doubt on the copy cat material scenario. It was reported that some bad imitations of C-276 contained higher amounts of carbon. Carbon has a very high affinity for combining with anything, chromium most of all. If our failed component was a cheap imitation then the precipitates should have contained higher levels of chromium in the form of chromium carbides, which they did not.



There was speculation that the precipitate could be an intermetallic phase. A closer look of the precipitates (4,000X) with EDX showed them to be rich in molybdenum, on the order of four times greater. The chromium level was equivalent to the reference sample.

What you are looking at is a "Mu phase" intermetallic. This intermetallic phase occurs when heated in a temperature range between 9000 - 1,5000F. Precipitates of this kind along the grain boundaries make the grain boundaries more susceptible to corrosion and less ductile. If the material was less ductile then the grains would have to show signs of being torn away as opposed to having a smooth abraded appearance under normal wear circumstances.




A normal wear pattern would tend to show an abraded surface with little defining topography. The magnitude of a typical crack is highlighted by the three red dots in the picture (1,000X). The grains were not being abraded. The topography of the area surrounding the intergranular crack in our failed nozzle shows signs of cleavage. The dimple shape is flat. Cleavage is a brittle phenomenon. In the case of the failed nozzle, less ductility meant that the high velocity of the emerging stream ripped the material away one grain at a time.



There were some indications of ductile failure. As shown in the picture (1,000X), the dimples are round and have depth to them.

As a side note about something that has nothing to do with the immediate failure analysis, note the inclusions contained within some of the dimples (near the red dots). The inclusions acted as nucleation sights for the crack initiation. In short, the "Big" crack is nothing more than a whole lot of little cracks forming together. Each round dimple is its own crack. The crack starts at an inclusion and radiates outward in a radial pattern, eventually joining up with other tiny cracks that started in much the same fashion.



The microstructure of the heat affected material shows significantly larger grains than the reference sample of C-276. The magnification of the bad sample, shown here at 100X, is half as great as that of the good sample, yet the grains appear to be twice the size. The reason for the difference in size is that the grains grew together in the presence of heat. According to the literature, this material has a history of occurrences where the material becomes unstable at high temperature and forms intermetallic phases. Therefore this material should never have been used for this service. A material such as C-4 should have been used. It has better thermal stability in the sensitizing temperature range mentioned above.

In light of this information the C-276 material still performed in an outstanding manner over a 13 year period. Further investigation revealed that there had been occurances in the past where the nozzle wore out prematurely. There was a distinct possibility that the flame coming from this nozzle was burning hotter than normal. There were some rare upset conditions that could cause this condition to happen.

However, there was no way to confirm this ... and nobody was talking.

The machine shop foreman lived.


All Pictures and Text Copyright © 2001 - 2016 Contact Mr. Adler Adler Engineering LLC of Wyoming USA

Great care has been taken in the compilation of this article. However, no warranty, expressed or implied, including without limitation, warranties of merchantability or fitness for a particular purpose, are given in connection with this article or any article archived on this website. Although this information is believed to be accurate by the author, the author cannot guarantee favorable results will be obtained from the use of this article alone. This article is intended for use by persons at their sole discretion and risk. Since the conditions of product or material use are outside of the author's control, the author assumes no liability or obligation in connection with any use of this information. The author is not liable for special, indirect or consequential damages resulting from the use of this material.

No part of this article or any article archived in this website, or any part thereof, may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning or otherwise, without the prior written permission of the copyright holder R. H. Adler. Nothing contained in this article or any article archived in this website shall be construed as a grant of any right of manufacture, sale, use, or reproduction, in connection with any method, process, apparatus, product, composition, or system, whether or not covered by letters of patent, copyright, or trademark, and nothing contained in this article or any article archived in this website, shall be construed as a defense against any alleged infringement of letters of patent, copyright, or trademark, or as a defense against any liability for such infringement.