The following is a letter sent to Dr. Whitney in December. To date we have received no response.

Dr. E. Dow Whitney
University of Florida
Department of Materials Science and Engineering
165 Rhines Hall
Gainsville, FL 32611–2066
11 December 1999
Dr. Whitney:

We are writing you in reference to your November 1990 publication entitled “A Study of Marcite (Plaster) Deterioration in Swimming Pools,” which has been distributed to the pool industry in draft form.
The swimming pool plaster (marcite) phenomenon referred to as “spot etching” has continued to plague our industry since the release of your study. As part of an ongoing effort to identify the cause of this elusive defect, many references have been made to your study. However, we have a number of questions relative to the study, which we hope you can clarify for us.

Definition
Part of the confusion about spot etching exists because of a lack of distinction by some between common, general etching and the specific phenomenon of spot etching. General etching refers to a uniform roughness of the entire surface of a plaster, caused by a removal of cement paste, thus exposing rough edges of aggregate. Spot etching refers to round spots, usually 1/4 to 1/2 inch in diameter, with intact or smooth plaster surrounding the spots. Aggregate present in the spotted areas are smooth to the touch – indeed, spot etched plaster cannot be identified by touch alone. A possible third condition may be included, in that with some spot–patterned occurrences the claim is that an actual recessed pit is developed, in which the plaster becomes very soft (the word “putty” is used...) and aggregate falls out of the pit. For lack of a uniform term, perhaps we can refer to this type as soft spot pitting. The last paragraph of page 3 (continuing to page 4) in your study indicates that you are aware of the spot patterning that has plagued the industry. We are familiar with aggressive deterioration of plaster, and study on your part relative to such general etching may be interesting but does not explain the spot etching phenomenon. A definition of spot etching must, by nature, include a rationale of why certain defined spots manifest the effect, when the rest (usually the majority) of the plaster does not.
Although page 3 of your study indicates an understanding of this difference, none of the conclusions seem to differentiate between general etching and spot etching, or explain why spots happen rather than general. Indeed, in your Concrete International article (“Etching, Mottling, and Staining in Plaster Coated Concrete Swimming Pools”, by Cal Eden and E. Dow Whitney, September 1990, pages 31 and 32) you present a definition of etching that describes both general and spot patterns, but photographs of plaster that are only of the general pattern, sequentially described as “severely etched,” “heavy etching,” and “severe etching.” Again – although this study does define spot etching, it does not explain the spotting phenomenon.
Question #1 – Would you please comment on whether or not your lab study and/or your examinations of field plaster samples specifically relate to spot etching, as opposed to general plaster etching or deterioration?

Plaster Chemistry
Another problem with the study is that it is based on fundamental misunderstandings relative to cement chemistry as it applies to the pool environment. One example of this flaw is a misunderstanding of the source of plaster dust.
On pages 31 and 32 of your study the plaster dust topic is addressed. The statement is made that adding basic water to a pool will convert bicarbonate ion to carbonate ion. This process is said to be the source of “plaster dust” in swimming pools. It is specifically noted that none of the plaster dust in new plaster pools comes from the cement itself. However many, if not most in the industry would disagree with this statement. Conventional wisdom is that calcium hydroxide is a normal reaction product of hydrating cement, and it is a portion of this hydroxide product that is released into the pool water and forms a calcium carbonate precipitate in most pools by reacting with bicarbonate in the pool water.
Question #2 – In light of your departure from normally accepted theory relative to plaster dust formation, would you please comment on your theory?

References
Crucial material that is extensively quoted without reference in this study previously appeared in other work dating to the early 1970s. Specifically, the following paragraphs have appeared elsewhere previously, virtually verbatim:

Page 28 Section 5.1 1st sentence
Page 29 Section 5.1 Last sentence of 1st paragraph and entire final paragraph
Page 29–31 Section 5.2 The entire page and a half of Section 5.2 except for the first sentence (with a few very minor wording or formulaic alterations)
Page 31–32 Section 5.3 Last paragraph under formula (5), Formulas (6) and (7), and the
following 2 sentences

Among other things, the questioned description of plaster dust formation appears in this earlier document.
Question #3 – Would you comment on the source of these unreferenced passages?

Calcium Hydroxide and the Surface of Pool Plaster
In your study the presence of calcium hydroxide in swimming pool plaster is repeatedly affirmed. However, although calcium hydroxide is a normal phase of cement, it is our experience, based on an understanding of the pool environment, and reinforced by the analysis of various cured pool plaster samples, that calcium hydroxide is not a significant component of the top few millimeters of cured pool plaster. Once the majority of the curing is complete (ca. 2–3 weeks after filling with water), hydroxide on the surface has either been converted chemically to carbonates or other more durable compounds in the bulk cement paste, or solubilized in the pool water. Indeed, water balanced in the bicarbonate range (i.e., <8.3) should immediately perform this chemical reaction with hydroxide present in any portion of the plaster which has contact with the water. It is normal, however, to see expected amounts of hydroxide in the interior (i.e., deeper than a few millimeters) portion of the bulk cement paste.
On pages 14–16 of your study, it is reported that a calcium hydroxide phase was present in non–etched portions of pool plaster samples. This determination is reported to have been made via X–ray diffraction. Unfortunately, none of the actual documentation – pictures, X–ray diffraction analysis, SEM scans, or even comprehensive descriptions are given. It is therefore impossible to verify the claims made in the report.
Figure 3 on page 21 is used to show that, according to EDS profiling, the percentage of the surface composed of calcium decrease from about 94.5% in non–etched areas to about 84.5% in etched areas. However, this data does not seem to differentiate between specific phases of calcium. Our analysis of similar samples has led us to believe that a hydroxide phase usually would not exist in either the etched or non–etched areas of spot etched plaster.
Question #4 – Would you please confirm and explain the presence of calcium hydroxide on the surface of underwater–cured plaster?
Question #5 – Would you please make available to us and the rest of the industry the scans, graphs, pictures, and descriptions which must have been generated as part of this study?

The Effects of Trowelling
We are unclear on what your ideas might be relative to the significance of trowelling. Although the coupon study did not examine trowelled, but rather mold–formed plaster, this section seems to indicate that you feel excess trowelling, or the bringing to the surface of excess calcium hydroxide is detrimental.
Question #6 – Was your intent to implicate excess trowelling which forces aggregate lower into the body of the plaster coating, thus creating a layer that is almost entirely cement paste? Do you have reason to believe that this process is directly related to spot etching?

The Lab Experiment
A centerpiece of your study seems to be the lab experiment wherein special coupons were exposed to a specific water, and weight analysis was reportedly performed. We have used your description of the experiment to recreate it, resulting in many questions and dissimilar conclusions. Would you please help clear up the following questions?
Question #7 – In order to create plaster coupons of the same composition as yours, we had difficulty ascertaining the ratio of cement to aggregate in the coupons. Indeed, since the description of the other components is specific, but an aggregate component is not described, it seems that your coupons were cement paste only. Was this the case? If so, why? The plaster surface to submersion water volume ratio, the circulation rate, etc. were all designed to mimic the pool environment, yet the actual product tested seems to have been a cement paste instead of pool plaster...
Question #8 – Your method of evaluation was weight, but no weight loss data is given. In our experience, the coupons actually increased in weight for a time as water was incorporated into the curing/hydrating material. On pages 24–26 of your report graphs are given which record activity, not by weight, but by “% reacted times 103 ”. Would you share the actual weight data, as well as a description of how the “% reacted times 103 ” correlates with that weight change? Also, was the calcium content of the submersion water determined at the end of the testing period, and if so was all calcium carbonate precipitate first dissolved? What were the results?
Question #9 – What determinations were made to suggest that formed plaster surfaces react in a significantly similar manner as trowelled surfaces?
Question #10 – Why do we have no definitive statement on whether the coupons spot etched, since the report includes a definitive definition of spot etching? Where are descriptions or photographs to substantiate the inference that they did? When we perform this experiment (with either aggregate–containing or aggregate free coupons), the coupons always etch evenly over the entire coupon, and never in a spot pattern.
Question #11 – What is the “percent reacted (x103)” of coupons in “balanced” water? In other words, what is the control value, the expected amount of calcium loss?
Question #12 – Why does the study infer that balancing the pH, alkalinity, and calcium levels of pool water are sufficient to protect the plaster from deterioration? Don’t the negative Gibbs free energy exchange values presented for calcium hydroxide predict that any water with a pH below 13 is, to varying degrees, aggressive to the calcium hydroxide in plaster?
Question #13 – Why does the study repeatedly refer to the pH range of 10–10.5 as the “natural pH” of a water solution exposed to plaster? Isn’t it because any water with a pH <13 in a plaster vessel will either convert calcium hydroxide to calcium carbonate in the bulk plaster, or else dissolve calcium hydroxide into calcium bicarbonate in the water, thus raising the water’s pH, and that the only way to stop this is not pool industry–defined water balance, but rather the maintenance of pH in the pool greater than 13?
Question #14 – What is actually your understanding of the term “preferentially leached” when referring to calcium hydroxide being removed from a plaster surface? It is mentioned in the report that “calcium hydroxide comprises about 25% by weight of the plaster”, and it can be assumed that the distribution of this calcium is relatively uniform throughout the surface after trowelling. No inference seems to be made in the study that calcium hydroxide is concentrated in round spots of the plaster surface, and that as a result any leaching of calcium hydroxide from the plaster comes only from these spots. However, no other explanation is given as to why a preferential leaching occurs on localized spots, rather than uniformly across a surface which should be subject to relatively uniform water chemistry. Is there therefore an impression that some spots are constitutionally more susceptible to etching?
Question #15 – Why, on page 32 of the study, is there the implication that gas chlorine is a cause of acidic water chemistry, but no such similar references are made to trichlor, muriatic acid, sodium bisulfate, or any other contributor to low pH? Isn’t this singling out of the misapplication of one product (i.e., gas chlorine in inadequately buffered water) and coupling it with an unsubstantiated capability of “selective solvation” from random spots of the surface inappropriate in such a scientific study?
Question #16 – What was the purpose of testing coupons in water with high concentrations of chlorine or cyanuric acid? Testing soft water (such as in the lab experiment), or chlorine and cyanuric acid (all water chemistry issues) without also examining any application issues (such as excessive retempering, hard trowelling, excessive calcium chloride, insufficient mix times, etc. etc.) gives at least the appearance of bias. At the time of the study, there were also some who suspected that there could be a materials contamination issue involved. Is there a reason why application and composition issues were ignored?

In light of these questions, we feel that the conclusions made in the study extend beyond the limits of the data, and that they should not be given weight until you can provide meaningful answers to these questions. We eagerly await your reply. We can be reached via the following methods:

Mail: Que Hales Phone: 520–573–6696
3114 E Pennsylvania St Fax: 520–625–1918
Tucson, AZ 85714 Email: que@poolchlor.com

Sincerely,
Kim Skinner – President, Pool Chlor
Que Hales – Manager, Pool Chlor of Tucson
Doug Latta – President, Aqua Clear